idnits 2.17.1 draft-ietf-dnsop-session-signal-10.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 the creation date from RFC1035, updated by this document, for RFC5378 checks: 1987-11-01) -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (June 07, 2018) is 2148 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) == Missing Reference: 'TBA1' is mentioned on line 1920, but not defined == Missing Reference: 'TBA2' is mentioned on line 1926, but not defined == Outdated reference: A later version (-04) exists of draft-ietf-dnssd-mdns-relay-00 == Outdated reference: A later version (-25) exists of draft-ietf-dnssd-push-14 == Outdated reference: A later version (-06) exists of draft-ietf-dprive-padding-policy-05 Summary: 0 errors (**), 0 flaws (~~), 6 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 DNSOP Working Group R. Bellis 3 Internet-Draft ISC 4 Updates: 1035, 7766 (if approved) S. Cheshire 5 Intended status: Standards Track Apple Inc. 6 Expires: December 9, 2018 J. Dickinson 7 S. Dickinson 8 Sinodun 9 T. Lemon 10 Barefoot Consulting 11 T. Pusateri 12 Unaffiliated 13 June 07, 2018 15 DNS Stateful Operations 16 draft-ietf-dnsop-session-signal-10 18 Abstract 20 This document defines a new DNS OPCODE for DNS Stateful Operations 21 (DSO). DSO messages communicate operations within persistent 22 stateful sessions, using type-length-value (TLV) syntax. Three TLVs 23 are defined that manage session timeouts, termination, and encryption 24 padding, and a framework is defined for extensions to enable new 25 stateful operations. This document updates RFC 1035 by adding a new 26 DNS header opcode and result code which has different message 27 semantics. This document updates RFC 7766 by redefining a session, 28 providing new guidance on connection re-use, and providing a new 29 mechanism for handling session idle timeouts. 31 Status of This Memo 33 This Internet-Draft is submitted in full conformance with the 34 provisions of BCP 78 and BCP 79. 36 Internet-Drafts are working documents of the Internet Engineering 37 Task Force (IETF). Note that other groups may also distribute 38 working documents as Internet-Drafts. The list of current Internet- 39 Drafts is at https://datatracker.ietf.org/drafts/current/. 41 Internet-Drafts are draft documents valid for a maximum of six months 42 and may be updated, replaced, or obsoleted by other documents at any 43 time. It is inappropriate to use Internet-Drafts as reference 44 material or to cite them other than as "work in progress." 46 This Internet-Draft will expire on December 9, 2018. 48 Copyright Notice 50 Copyright (c) 2018 IETF Trust and the persons identified as the 51 document authors. All rights reserved. 53 This document is subject to BCP 78 and the IETF Trust's Legal 54 Provisions Relating to IETF Documents 55 (https://trustee.ietf.org/license-info) in effect on the date of 56 publication of this document. Please review these documents 57 carefully, as they describe your rights and restrictions with respect 58 to this document. Code Components extracted from this document must 59 include Simplified BSD License text as described in Section 4.e of 60 the Trust Legal Provisions and are provided without warranty as 61 described in the Simplified BSD License. 63 Table of Contents 65 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 66 2. Requirements Language . . . . . . . . . . . . . . . . . . . . 5 67 3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 68 4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 9 69 5. Protocol Details . . . . . . . . . . . . . . . . . . . . . . 10 70 5.1. DSO Session Establishment . . . . . . . . . . . . . . . . 10 71 5.1.1. Connection Sharing . . . . . . . . . . . . . . . . . 12 72 5.1.2. Zero Round-Trip Operation . . . . . . . . . . . . . . 12 73 5.1.3. Middlebox Considerations . . . . . . . . . . . . . . 13 74 5.2. Message Format . . . . . . . . . . . . . . . . . . . . . 14 75 5.2.1. DNS Header Fields in DSO Messages . . . . . . . . . . 15 76 5.2.2. DSO Data . . . . . . . . . . . . . . . . . . . . . . 17 77 5.2.3. EDNS(0) and TSIG . . . . . . . . . . . . . . . . . . 22 78 5.3. Message Handling . . . . . . . . . . . . . . . . . . . . 23 79 5.3.1. Error Responses . . . . . . . . . . . . . . . . . . . 24 80 5.4. DSO Response Generation . . . . . . . . . . . . . . . . . 25 81 5.5. Responder-Initiated Operation Cancellation . . . . . . . 26 82 6. DSO Session Lifecycle and Timers . . . . . . . . . . . . . . 27 83 6.1. DSO Session Initiation . . . . . . . . . . . . . . . . . 27 84 6.2. DSO Session Timeouts . . . . . . . . . . . . . . . . . . 27 85 6.3. Inactive DSO Sessions . . . . . . . . . . . . . . . . . . 28 86 6.4. The Inactivity Timeout . . . . . . . . . . . . . . . . . 29 87 6.4.1. Closing Inactive DSO Sessions . . . . . . . . . . . . 29 88 6.4.2. Values for the Inactivity Timeout . . . . . . . . . . 30 89 6.5. The Keepalive Interval . . . . . . . . . . . . . . . . . 31 90 6.5.1. Keepalive Interval Expiry . . . . . . . . . . . . . . 31 91 6.5.2. Values for the Keepalive Interval . . . . . . . . . . 31 92 6.6. Server-Initiated Session Termination . . . . . . . . . . 33 93 6.6.1. Server-Initiated Retry Delay Message . . . . . . . . 34 94 7. Base TLVs for DNS Stateful Operations . . . . . . . . . . . . 37 95 7.1. Keepalive TLV . . . . . . . . . . . . . . . . . . . . . . 37 96 7.1.1. Client handling of received Session Timeout values . 39 97 7.1.2. Relation to EDNS(0) TCP Keepalive Option . . . . . . 41 98 7.2. Retry Delay TLV . . . . . . . . . . . . . . . . . . . . . 42 99 7.2.1. Retry Delay TLV used as a Primary TLV . . . . . . . . 42 100 7.2.2. Retry Delay TLV used as a Response Additional TLV . . 44 101 7.3. Encryption Padding TLV . . . . . . . . . . . . . . . . . 44 102 8. Summary Highlights . . . . . . . . . . . . . . . . . . . . . 45 103 8.1. QR bit and MESSAGE ID . . . . . . . . . . . . . . . . . . 45 104 8.2. TLV Usage . . . . . . . . . . . . . . . . . . . . . . . . 46 105 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 48 106 9.1. DSO OPCODE Registration . . . . . . . . . . . . . . . . . 48 107 9.2. DSO RCODE Registration . . . . . . . . . . . . . . . . . 48 108 9.3. DSO Type Code Registry . . . . . . . . . . . . . . . . . 48 109 10. Security Considerations . . . . . . . . . . . . . . . . . . . 49 110 11. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 49 111 12. References . . . . . . . . . . . . . . . . . . . . . . . . . 50 112 12.1. Normative References . . . . . . . . . . . . . . . . . . 50 113 12.2. Informative References . . . . . . . . . . . . . . . . . 51 114 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 52 116 1. Introduction 118 The use of transports for DNS other than UDP is being increasingly 119 specified, for example, DNS over TCP [RFC1035] [RFC7766] and DNS over 120 TLS [RFC7858]. Such transports can offer persistent, long-lived 121 sessions and therefore when using them for transporting DNS messages 122 it is of benefit to have a mechanism that can establish parameters 123 associated with those sessions, such as timeouts. In such situations 124 it is also advantageous to support server-initiated messages. 126 The existing EDNS(0) Extension Mechanism for DNS [RFC6891] is 127 explicitly defined to only have "per-message" semantics. While 128 EDNS(0) has been used to signal at least one session-related 129 parameter (the EDNS(0) TCP Keepalive option [RFC7828]) the result is 130 less than optimal due to the restrictions imposed by the EDNS(0) 131 semantics and the lack of server-initiated signalling. For example, 132 a server cannot arbitrarily instruct a client to close a connection 133 because the server can only send EDNS(0) options in responses to 134 queries that contained EDNS(0) options. 136 This document defines a new DNS OPCODE, DSO ([TBA1], tentatively 6), 137 for DNS Stateful Operations. DSO messages are used to communicate 138 operations within persistent stateful sessions, expressed using type- 139 length-value (TLV) syntax. This document defines an initial set of 140 three TLVs, used to manage session timeouts, termination, and 141 encryption padding. 143 The three TLVs defined here are all mandatory for all implementations 144 of DSO. Further TLVs may be defined in additional specifications. 146 The format for DSO messages (Section 5.2) differs somewhat from the 147 traditional DNS message format used for standard queries and 148 responses. The standard twelve-byte header is used, but the four 149 count fields (QDCOUNT, ANCOUNT, NSCOUNT, ARCOUNT) are set to zero and 150 accordingly their corresponding sections are not present. The actual 151 data pertaining to DNS Stateful Operations (expressed in TLV syntax) 152 is appended to the end of the DNS message header. When displayed 153 using packet analyzer tools that have not been updated to recognize 154 the DSO format, this will result in the DSO data being displayed as 155 unknown additional data after the end of the DNS message. It is 156 likely that future updates to these tools will add the ability to 157 recognize, decode, and display the DSO data. 159 This new format has distinct advantages over an RR-based format 160 because it is more explicit and more compact. Each TLV definition is 161 specific to its use case, and as a result contains no redundant or 162 overloaded fields. Importantly, it completely avoids conflating DNS 163 Stateful Operations in any way with normal DNS operations or with 164 existing EDNS(0)-based functionality. A goal of this approach is to 165 avoid the operational issues that have befallen EDNS(0), particularly 166 relating to middlebox behaviour. 168 With EDNS(0), multiple options may be packed into a single OPT 169 pseudo-RR, and there is no generalized mechanism for a client to be 170 able to tell whether a server has processed or otherwise acted upon 171 each individual option within the combined OPT pseudo-RR. The 172 specifications for each individual option need to define how each 173 different option is to be acknowledged, if necessary. 175 In contrast to EDNS(0), with DSO there is no compelling motivation to 176 pack multiple operations into a single message for efficiency 177 reasons, because DSO always operates using a connection-oriented 178 transport protocol. Each DSO operation is communicated in its own 179 separate DNS message, and the transport protocol can take care of 180 packing several DNS messages into a single IP packet if appropriate. 181 For example, TCP can pack multiple small DNS messages into a single 182 TCP segment. This simplification allows for clearer semantics. Each 183 DSO request message communicates just one primary operation, and the 184 RCODE in the corresponding response message indicates the success or 185 failure of that operation. 187 2. Requirements Language 189 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 190 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 191 "OPTIONAL" in this document are to be interpreted as described in BCP 192 14 [RFC2119] [RFC8174] when, and only when, they appear in all 193 capitals, as shown here. 195 3. Terminology 197 "DSO" is used to mean DNS Stateful Operation. 199 The term "connection" means a bidirectional byte (or message) stream, 200 where the bytes (or messages) are delivered reliably and in-order, 201 such as provided by using DNS over TCP [RFC1035] [RFC7766] or DNS 202 over TLS [RFC7858]. 204 The unqualified term "session" in the context of this document means 205 the exchange of DNS messages over a connection where: 207 o The connection between client and server is persistent and 208 relatively long-lived (i.e., minutes or hours, rather than 209 seconds). 211 o Either end of the connection may initiate messages to the other. 213 In this document the term "session" is used exclusively as described 214 above. The term has no relationship to the "session layer" of the 215 OSI "seven-layer model". 217 A "DSO Session" is established between two endpoints that acknowledge 218 persistent DNS state via the exchange of DSO messages over the 219 connection. This is distinct from a DNS-over-TCP session as 220 described in the previous specification for DNS over TCP [RFC7766]. 222 A "DSO Session" is terminated when the underlying connection is 223 closed. The underlying connection can be closed in two ways: 225 Where this specification says, "close gracefully," that means sending 226 a TLS close_notify (if TLS is in use) followed by a TCP FIN, or the 227 equivalents for other protocols. Where this specification requires a 228 connection to be closed gracefully, the requirement to initiate that 229 graceful close is placed on the client, to place the burden of TCP's 230 TIME-WAIT state on the client rather than the server. 232 Where this specification says, "forcibly abort," that means sending a 233 TCP RST, or the equivalent for other protocols. In the BSD Sockets 234 API this is achieved by setting the SO_LINGER option to zero before 235 closing the socket. 237 The term "server" means the software with a listening socket, 238 awaiting incoming connection requests. 240 The term "client" means the software which initiates a connection to 241 the server's listening socket. 243 The terms "initiator" and "responder" correspond respectively to the 244 initial sender and subsequent receiver of a DSO request message or 245 unacknowledged message, regardless of which was the "client" and 246 "server" in the usual DNS sense. 248 The term "sender" may apply to either an initiator (when sending a 249 DSO request message or unacknowledged message) or a responder (when 250 sending a DSO response message). 252 Likewise, the term "receiver" may apply to either a responder (when 253 receiving a DSO request message or unacknowledged message) or an 254 initiator (when receiving a DSO response message). 256 In protocol implementation there are generally two kinds of errors 257 that software writers have to deal with. The first is situations 258 that arise due to factors in the environment, such as temporary loss 259 of connectivity. While undesirable, these situations do not indicate 260 a flaw in the software, and they are situations that software should 261 generally be able to recover from. The second is situations that 262 should never happen when communicating with a correctly-implemented 263 peer. If they do happen, they indicate a serious flaw in the 264 protocol implementation, beyond what it is reasonable to expect 265 software to recover from. This document describes this latter form 266 of error condition as a "fatal error" and specifies that an 267 implementation encountering a fatal error condition "MUST forcibly 268 abort the connection immediately". Given that these fatal error 269 conditions signify defective software, and given that defective 270 software is likely to remain defective for some time until it is 271 fixed, after forcibly aborting a connection, a client SHOULD refrain 272 from automatically reconnecting to that same service instance for at 273 least one hour. 275 This document uses the term "same service instance" as follows: 277 o In cases where a server is specified or configured using an IP 278 address and TCP port number, two different configurations are 279 referring to the same service instance if they contain the same IP 280 address and TCP port number. 282 o In cases where a server is specified or configured using a 283 hostname and TCP port number, such as in the content of a DNS SRV 284 record [RFC2782], two different configurations (or DNS SRV 285 records) are considered to be referring to the same service 286 instance if they contain the same hostname (subject to the usual 287 case insensitive DNS name matching rules [RFC1034] [RFC1035]) and 288 TCP port number. In these cases, configurations with different 289 hostnames are considered to be referring to different service 290 instances, even if those different hostnames happen to be aliases, 291 or happen to resolve to the same IP address(es). Implementations 292 SHOULD NOT resolve hostnames and then perform matching of IP 293 address(es) in order to evaluate whether two entities should be 294 determined to be the "same service instance". 296 When an anycast service is configured on a particular IP address and 297 port, it must be the case that although there is more than one 298 physical server responding on that IP address, each such server can 299 be treated as equivalent. If a change in network topology causes 300 packets in a particular TCP connection to be sent to an anycast 301 server instance that does not know about the connection, the normal 302 keepalive and TCP connection timeout process will allow for recovery. 303 If after the connection is, the client's assumption that it is 304 connected to the same service is violated in some way, that would be 305 considered to be incorrect behavior in this context. It is however 306 out of the possible scope for this specification to make specific 307 recommendations in this regard; that would be up to follow-on 308 documents that describe specific uses of DNS stateful operations. 310 The term "long-lived operations" refers to operations such as Push 311 Notification subscriptions [I-D.ietf-dnssd-push], Discovery Relay 312 interface subscriptions [I-D.ietf-dnssd-mdns-relay], and other future 313 long-lived DNS operations that choose to use DSO as their basis. 314 These operations establish state that persists beyond the lifetime of 315 a traditional brief request/response transaction. This document, the 316 base specification for DNS Stateful Operations, defines a framework 317 for supporting long-lived operations, but does not itself define any 318 long-lived operations. Nonetheless, to appreciate the design 319 rationale behind DNS Stateful Operations, it is helpful to understand 320 the kind of long-lived operations that it is intended to support. 322 DNS Stateful Operations uses three kinds of message: "DSO request 323 messages", "DSO response messages", and "DSO unacknowledged 324 messages". A DSO request message elicits a DSO response message. 325 DSO unacknowledged messages are unidirectional messages and do not 326 generate any response. 328 Both DSO request messages and DSO unacknowledged messages are 329 formatted as DNS request messages (the header QR bit is set to zero, 330 as described in Section 5.2). One difference is that in DSO request 331 messages the MESSAGE ID field is nonzero; in DSO unacknowledged 332 messages it is zero. 334 The content of DSO messages is expressed using type-length-value 335 (TLV) syntax. 337 In a DSO request message or DSO unacknowledged message the first TLV 338 is referred to as the "Primary TLV" and determines the nature of the 339 operation being performed, including whether it is an acknowledged or 340 unacknowledged operation; any other TLVs in a DSO request message or 341 unacknowledged message are referred to as "Additional TLVs" and serve 342 additional non-primary purposes, which may be related to the primary 343 purpose, or not, as in the case of the encryption padding TLV. 345 A DSO response message may contain no TLVs, or it may contain one or 346 more TLVs as appropriate to the information being communicated. In 347 the context of DSO response messages, one or more TLVs with the same 348 DSO-TYPE as the Primary TLV in the corresponding DSO request message 349 are referred to as "Response Primary TLVs". Any other TLVs with 350 different DSO-TYPEs are referred to as "Response Additional TLVs". 351 The Response Primary TLV(s), if present, MUST occur first in the 352 response message, before any Response Additional TLVs. 354 Two timers (elapsed time since an event) are defined in this 355 document: 357 o an inactivity timer (see Section 6.4 and Section 7.1) 359 o a keepalive timer (see Section 6.5 and Section 7.1) 361 The timeouts associated with these timers are called the inactivity 362 timeout and the keepalive interval, respectively. The term "Session 363 Timeouts" is used to refer to this pair of timeout values. 365 Resetting a timer means resetting the timer value to zero and 366 starting the timer again. Clearing a timer means resetting the timer 367 value to zero but NOT starting the timer again. 369 4. Discussion 371 There are several use cases for DNS Stateful operations that can be 372 described here. 374 Firstly, establishing session parameters such as server-defined 375 timeouts is of great use in the general management of persistent 376 connections. For example, using DSO sessions for stub-to-recursive 377 DNS-over-TLS [RFC7858] is more flexible for both the client and the 378 server than attempting to manage sessions using just the EDNS(0) TCP 379 Keepalive option [RFC7828]. The simple set of TLVs defined in this 380 document is sufficient to greatly enhance connection management for 381 this use case. 383 Secondly, DNS-SD [RFC6763] has evolved into a naturally session-based 384 mechanism where, for example, long-lived subscriptions lend 385 themselves to 'push' mechanisms as opposed to polling. Long-lived 386 stateful connections and server-initiated messages align with this 387 use case [I-D.ietf-dnssd-push]. 389 A general use case is that DNS traffic is often bursty but session 390 establishment can be expensive. One challenge with long-lived 391 connections is to maintain sufficient traffic to maintain NAT and 392 firewall state. To mitigate this issue this document introduces a 393 new concept for the DNS, that is DSO "Keepalive traffic". This 394 traffic carries no DNS data and is not considered 'activity' in the 395 classic DNS sense, but serves to maintain state in middleboxes, and 396 to assure client and server that they still have connectivity to each 397 other. 399 5. Protocol Details 401 5.1. DSO Session Establishment 403 DSO messages MUST be carried in only protocols and in environments 404 where a session may be established according to the definition given 405 above in the Terminology section (Section 3). 407 DNS over plain UDP [RFC0768] is not appropriate since it fails on the 408 requirement for in-order message delivery, and, in the presence of 409 NAT gateways and firewalls with short UDP timeouts, it fails to 410 provide a persistent bi-directional communication channel unless an 411 excessive amount of keepalive traffic is used. 413 At the time of publication, DSO is specified only for DNS over TCP 414 [RFC1035] [RFC7766], and for DNS over TLS over TCP [RFC7858]. Any 415 use of DSO over some other connection technology needs to be 416 specified in an appropriate future document. 418 Determining whether a given connection is using DNS over TCP, or DNS 419 over TLS over TCP, is outside the scope of this specification, and 420 must be determined using some out-of-band configuration information. 421 There is no provision within the DSO specification to turn TLS on or 422 off during the lifetime of a connection. For service types where the 423 service instance is discovered using a DNS SRV record [RFC2782], the 424 specification for that service type SRV name [RFC6335] will state 425 whether the connection uses plain TCP, or TLS over TCP. For example, 426 the specification for the "_dns-push-tls._tcp" service 427 [I-D.ietf-dnssd-push], states that it uses TLS. It is a common 428 convention that protocols specified to run over TLS are given IANA 429 service type names ending in "-tls". 431 In some environments it may be known in advance by external means 432 that both client and server support DSO, and in these cases either 433 client or server may initiate DSO messages at any time. 435 However, in the typical case a server will not know in advance 436 whether a client supports DSO, so in general, unless it is known in 437 advance by other means that a client does support DSO, a server MUST 438 NOT initiate DSO request messages or DSO unacknowledged messages 439 until a DSO Session has been mutually established by at least one 440 successful DSO request/response exchange initiated by the client, as 441 described below. Similarly, unless it is known in advance by other 442 means that a server does support DSO, a client MUST NOT initiate DSO 443 unacknowledged messages until after a DSO Session has been mutually 444 established. 446 A DSO Session is established over a connection by the client sending 447 a DSO request message, such as a DSO Keepalive request message 448 (Section 7.1), and receiving a response, with matching MESSAGE ID, 449 and RCODE set to NOERROR (0), indicating that the DSO request was 450 successful. 452 If the RCODE in the response is set to DSOTYPENI ("DSO-TYPE Not 453 Implemented", [TBA2] tentatively RCODE 11) this indicates that the 454 server does support DSO, but does not implement the DSO-TYPE of the 455 primary TLV in this DSO request message. A server implementing DSO 456 MUST NOT return DSOTYPENI for a DSO Keepalive request message, 457 because the Keepalive TLV is mandatory to implement. But in the 458 future, if a client attempts to establish a DSO Session using a 459 response-requiring DSO request message using some newly-defined DSO- 460 TYPE that the server does not understand, that would result in a 461 DSOTYPENI response. If the server returns DSOTYPENI then a DSO 462 Session is not considered established, but the client is permitted to 463 continue sending DNS messages on the connection, including other DSO 464 messages such as the DSO Keepalive, which may result in a successful 465 NOERROR response, yielding the establishment of a DSO Session. 467 If the RCODE is set to any value other than NOERROR (0) or DSOTYPENI 468 ([TBA2] tentatively 11), then the client MUST assume that the server 469 does not implement DSO at all. In this case the client is permitted 470 to continue sending DNS messages on that connection, but the client 471 SHOULD NOT issue further DSO messages on that connection. 473 When the server receives a DSO request message from a client, and 474 transmits a successful NOERROR response to that request, the server 475 considers the DSO Session established. 477 When the client receives the server's NOERROR response to its DSO 478 request message, the client considers the DSO Session established. 480 Once a DSO Session has been established, either end may unilaterally 481 send appropriate DSO messages at any time, and therefore either 482 client or server may be the initiator of a message. 484 Once a DSO Session has been established, clients and servers should 485 behave as described in this specification with regard to inactivity 486 timeouts and session termination, not as previously prescribed in the 487 earlier specification for DNS over TCP [RFC7766]. 489 Note that for clients that implement only the DSO-TYPEs defined in 490 this base specification, sending a DSO Keepalive TLV is the only DSO 491 request message they have available to initiate a DSO Session. Even 492 for clients that do implement other future DSO-TYPEs, for simplicity 493 they MAY elect to always send an initial DSO Keepalive request 494 message as their way of initiating a DSO Session. A future 495 definition of a new response-requiring DSO-TYPE gives implementers 496 the option of using that new DSO-TYPE if they wish, but does not 497 change the fact that sending a DSO Keepalive TLV remains a valid way 498 of initiating a DSO Session. 500 5.1.1. Connection Sharing 502 As previously specified for DNS over TCP [RFC7766]: 504 To mitigate the risk of unintentional server overload, DNS 505 clients MUST take care to minimize the number of concurrent 506 TCP connections made to any individual server. It is RECOMMENDED 507 that for any given client/server interaction there SHOULD be 508 no more than one connection for regular queries, one for zone 509 transfers, and one for each protocol that is being used on top 510 of TCP (for example, if the resolver was using TLS). However, 511 it is noted that certain primary/secondary configurations 512 with many busy zones might need to use more than one TCP 513 connection for zone transfers for operational reasons (for 514 example, to support concurrent transfers of multiple zones). 516 A single server may support multiple services, including DNS Updates 517 [RFC2136], DNS Push Notifications [I-D.ietf-dnssd-push], and other 518 services, for one or more DNS zones. When a client discovers that 519 the target server for several different operations is the same target 520 hostname and port, the client SHOULD use a single shared DSO Session 521 for all those operations. A client SHOULD NOT open multiple 522 connections to the same target host and port just because the names 523 being operated on are different or happen to fall within different 524 zones. This requirement is to reduce unnecessary connection load on 525 the DNS server. 527 However, server implementers and operators should be aware that 528 connection sharing may not be possible in all cases. A single host 529 device may be home to multiple independent client software instances 530 that don't coordinate with each other. Similarly, multiple 531 independent client devices behind the same NAT gateway will also 532 typically appear to the DNS server as different source ports on the 533 same client IP address. Because of these constraints, a DNS server 534 MUST be prepared to accept multiple connections from different source 535 ports on the same client IP address. 537 5.1.2. Zero Round-Trip Operation 539 There is increased awareness today of the performance benefits of 540 eliminating round trips in session establishment. Technologies like 541 TCP Fast Open [RFC7413] and TLS 1.3 [I-D.ietf-tls-tls13] provide 542 mechanisms to reduce or eliminate round trips in session 543 establishment. 545 Similarly, DSO supports zero round-trip operation. 547 Having initiated a connection to a server, possibly using zero round- 548 trip TCP Fast Open and/or zero round-trip TLS 1.3, a client MAY send 549 multiple response-requiring DSO request messages to the server in 550 succession without having to wait for a response to the first request 551 message to confirm successful establishment of a DSO session. 553 However, a client MUST NOT send non-response-requiring DSO request 554 messages until after a DSO Session has been mutually established. 556 Similarly, a server MUST NOT send DSO request messages until it has 557 received a response-requiring DSO request message from a client and 558 transmitted a successful NOERROR response for that request. 560 Caution must be taken to ensure that DSO messages sent before the 561 first round-trip is completed are idempotent, or are otherwise immune 562 to any problems that could be result from the inadvertent replay that 563 can occur with zero round-trip operation. 565 5.1.3. Middlebox Considerations 567 Where an application-layer middlebox (e.g., a DNS proxy, forwarder, 568 or session multiplexer) is in the path, the middlebox MUST NOT 569 blindly forward DSO messages in either direction, and MUST treat the 570 inbound and outbound connections as separate sessions. This does not 571 preclude the use of DSO messages in the presence of an IP-layer 572 middlebox, such as a NAT that rewrites IP-layer and/or transport- 573 layer headers but otherwise preserves the effect of a single session 574 between the client and the server. 576 To illustrate the above, consider a network where a middlebox 577 terminates one or more TCP connections from clients and multiplexes 578 the queries therein over a single TCP connection to an upstream 579 server. The DSO messages and any associated state are specific to 580 the individual TCP connections. A DSO-aware middlebox MAY in some 581 circumstances be able to retain associated state and pass it between 582 the client and server (or vice versa) but this would be highly TLV- 583 specific. For example, the middlebox may be able to maintain a list 584 of which clients have made Push Notification subscriptions 585 [I-D.ietf-dnssd-push] and make its own subscription(s) on their 586 behalf, relaying any subsequent notifications to the client (or 587 clients) that have subscribed to that particular notification. 589 5.2. Message Format 591 A DSO message begins with the standard twelve-byte DNS message header 592 [RFC1035] with the OPCODE field set to the DSO OPCODE ([TBA1] 593 tentatively 6). However, unlike standard DNS messages, the question 594 section, answer section, authority records section and additional 595 records sections are not present. The corresponding count fields 596 (QDCOUNT, ANCOUNT, NSCOUNT, ARCOUNT) MUST be set to zero on 597 transmission. 599 If a DSO message is received where any of the count fields are not 600 zero, then a FORMERR MUST be returned, unless a future IETF Standard 601 specifies otherwise. 603 1 1 1 1 1 1 604 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 605 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 606 | MESSAGE ID | 607 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 608 |QR | OPCODE | Z | RCODE | 609 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 610 | QDCOUNT (MUST be zero) | 611 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 612 | ANCOUNT (MUST be zero) | 613 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 614 | NSCOUNT (MUST be zero) | 615 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 616 | ARCOUNT (MUST be zero) | 617 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 618 | | 619 / DSO Data / 620 / / 621 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 623 5.2.1. DNS Header Fields in DSO Messages 625 In an unacknowledged message the MESSAGE ID field MUST be set to 626 zero. In an acknowledged request message the MESSAGE ID field MUST 627 be set to a unique nonzero value, that the initiator is not currently 628 using for any other active operation on this connection. For the 629 purposes here, a MESSAGE ID is in use in this DSO Session if the 630 initiator has used it in a request for which it is still awaiting a 631 response, or if the client has used it to set up a long-lived 632 operation that has not yet been cancelled. For example, a long-lived 633 operation could be a Push Notification subscription 634 [I-D.ietf-dnssd-push] or a Discovery Relay interface subscription 635 [I-D.ietf-dnssd-mdns-relay]. 637 Whether a message is acknowledged or unacknowledged is determined 638 only by the specification for the Primary TLV. An acknowledgment 639 cannot be requested by including a nonzero message ID in a message 640 the primary TLV of which is specified to be unacknowledged, nor can 641 an acknowledgment be prevented by sending a message ID of zero in a 642 message with a primary TLV that is specified to be acknowledged. A 643 responder that receives either such malformed message MUST treat it 644 as a fatal error and forcibly abort the connection immediately. 646 In a request or unacknowledged message the DNS Header QR bit MUST be 647 zero (QR=0). If the QR bit is not zero the message is not a request 648 or unacknowledged message. 650 In a response message the DNS Header QR bit MUST be one (QR=1). 651 If the QR bit is not one the message is not a response message. 653 In a response message (QR=1) the MESSAGE ID field MUST contain a copy 654 of the value of the MESSAGE ID field in the request message being 655 responded to. In a response message (QR=1) the MESSAGE ID field MUST 656 NOT be zero. If a response message (QR=1) is received where the 657 MESSAGE ID is zero this is a fatal error and the recipient MUST 658 forcibly abort the connection immediately. 660 The DNS Header OPCODE field holds the DSO OPCODE value ([TBA1] 661 tentatively 6). 663 The Z bits are currently unused in DSO messages, and in both DSO 664 requests and DSO responses the Z bits MUST be set to zero (0) on 665 transmission and MUST be silently ignored on reception, unless a 666 future IETF Standard specifies otherwise. 668 In a DNS request message (QR=0) the RCODE is set according to the 669 definition of the request. For example, in a Retry Delay message 670 (Section 6.6.1) the RCODE indicates the reason for termination. 671 However, in most cases, except where clearly specified otherwise, in 672 a DNS request message (QR=0) the RCODE is set to zero on 673 transmission, and silently ignored on reception. 675 The RCODE value in a response message (QR=1) may be one of the 676 following values: 678 +---------+-----------+---------------------------------------------+ 679 | Code | Mnemonic | Description | 680 +---------+-----------+---------------------------------------------+ 681 | 0 | NOERROR | Operation processed successfully | 682 | | | | 683 | 1 | FORMERR | Format error | 684 | | | | 685 | 2 | SERVFAIL | Server failed to process request due to a | 686 | | | problem with the server | 687 | | | | 688 | 3 | NXDOMAIN | Name Error -- Named entity does not exist | 689 | | | (TLV-dependent) | 690 | | | | 691 | 4 | NOTIMP | DSO not supported | 692 | | | | 693 | 5 | REFUSED | Operation declined for policy reasons | 694 | | | | 695 | 9 | NOTAUTH | Not Authoritative (TLV-dependent) | 696 | | | | 697 | [TBA2] | DSOTYPENI | Primary TLV's DSO-Type is not implemented | 698 | 11 | | | 699 +---------+-----------+---------------------------------------------+ 701 Use of the above RCODEs is likely to be common in DSO but does not 702 preclude the definition and use of other codes in future documents 703 that make use of DSO. 705 If a document defining a new DSO-TYPE makes use of NXDOMAIN (Name 706 Error) or NOTAUTH (Not Authoritative) then that document MUST specify 707 the specific interpretation of these RCODE values in the context of 708 that new DSO TLV. 710 5.2.2. DSO Data 712 The standard twelve-byte DNS message header with its zero-valued 713 count fields is followed by the DSO Data, expressed using TLV syntax, 714 as described below Section 5.2.2.1. 716 A DSO message may be a request message, a response message, or an 717 unacknowledged message. 719 A DSO request message or DSO unacknowledged message MUST contain at 720 least one TLV. The first TLV in a DSO request message or DSO 721 unacknowledged message is referred to as the "Primary TLV" and 722 determines the nature of the operation being performed, including 723 whether it is an acknowledged or unacknowledged operation. In some 724 cases it may be appropriate to include other TLVs in a request 725 message or unacknowledged message, such as the Encryption Padding TLV 726 (Section 7.3), and these extra TLVs are referred to as the 727 "Additional TLVs". 729 A DSO response message may contain no TLVs, or it may be specified to 730 contain one or more TLVs appropriate to the information being 731 communicated. 733 A DSO response message may contain one or more TLVs with DSO-TYPE the 734 same as the Primary TLV from the corresponding DSO request message, 735 in which case those TLV(s) are referred to as "Response Primary 736 TLVs". A DSO response message is not required to carry Response 737 Primary TLVs. The MESSAGE ID field in the DNS message header is 738 sufficient to identify the DSO request message to which this response 739 message relates. 741 A DSO response message may contain one or more TLVs with DSO-TYPEs 742 different from the Primary TLV from the corresponding DSO request 743 message, in which case those TLV(s) are referred to as "Response 744 Additional TLVs". 746 Response Primary TLV(s), if present, MUST occur first in the response 747 message, before any Response Additional TLVs. 749 It is anticipated that most DSO operations will be specified to use 750 request messages, which generate corresponding responses. In some 751 specialized high-traffic use cases, it may be appropriate to specify 752 unacknowledged messages. Unacknowledged messages can be more 753 efficient on the network, because they don't generate a stream of 754 corresponding reply messages. Using unacknowledged messages can also 755 simplify software in some cases, by removing need for an initiator to 756 maintain state while it waits to receive replies it doesn't care 757 about. When the specification for a particular TLV states that, when 758 used as a Primary TLV (i.e., first) in an outgoing DNS request 759 message (i.e., QR=0), that message is to be unacknowledged, the 760 MESSAGE ID field MUST be set to zero and the receiver MUST NOT 761 generate any response message corresponding to this unacknowledged 762 message. 764 The previous point, that the receiver MUST NOT generate responses to 765 unacknowledged messages, applies even in the case of errors. When a 766 DSO message is received where both the QR bit and the MESSAGE ID 767 field are zero, the receiver MUST NOT generate any response. For 768 example, if the DSO-TYPE in the Primary TLV is unrecognized, then a 769 DSOTYPENI error MUST NOT be returned; instead the receiver MUST 770 forcibly abort the connection immediately. 772 Unacknowledged messages MUST NOT be used "speculatively" in cases 773 where the sender doesn't know if the receiver supports the Primary 774 TLV in the message, because there is no way to receive any response 775 to indicate success or failure of the request message (the request 776 message does not contain a unique MESSAGE ID with which to associate 777 a response with its corresponding request). Unacknowledged messages 778 are only appropriate in cases where the sender already knows that the 779 receiver supports, and wishes to receive, these messages. 781 For example, after a client has subscribed for Push Notifications 782 [I-D.ietf-dnssd-push], the subsequent event notifications are then 783 sent as unacknowledged messages, and this is appropriate because the 784 client initiated the message stream by virtue of its Push 785 Notification subscription, thereby indicating its support of Push 786 Notifications, and its desire to receive those notifications. 788 Similarly, after an Discovery Relay client has subscribed to receive 789 inbound mDNS (multicast DNS, [RFC6762]) traffic from an Discovery 790 Relay, the subsequent stream of received packets is then sent using 791 unacknowledged messages, and this is appropriate because the client 792 initiated the message stream by virtue of its Discovery Relay link 793 subscription, thereby indicating its support of Discovery Relay, and 794 its desire to receive inbound mDNS packets over that DSO session 795 [I-D.ietf-dnssd-mdns-relay]. 797 5.2.2.1. TLV Syntax 799 All TLVs, whether used as "Primary", "Additional", "Response 800 Primary", or "Response Additional", use the same encoding syntax. 802 The specification for a TLV states whether that DSO-TYPE may be used 803 in "Primary", "Additional", "Response Primary", or "Response 804 Additional" TLVs. The specification for a TLV also states whether, 805 when used as the Primary (i.e., first) TLV in a DNS request message 806 (i.e., QR=0), that DSO message is to be acknowledged. If the DSO 807 message is to be acknowledged, the specification also states which 808 TLVs, if any, are to be included in the response. The Primary TLV 809 may or may not be contained in the response, depending on what is 810 stated in the specification for that TLV. 812 1 1 1 1 1 1 813 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 814 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 815 | DSO-TYPE | 816 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 817 | DSO-LENGTH | 818 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 819 | | 820 / DSO-DATA / 821 / / 822 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 824 DSO-TYPE: A 16-bit unsigned integer, in network (big endian) byte 825 order, giving the DSO-TYPE of the current DSO TLV per the IANA DSO 826 Type Code Registry. 828 DSO-LENGTH: A 16-bit unsigned integer, in network (big endian) byte 829 order, giving the size in bytes of the DSO-DATA. 831 DSO-DATA: Type-code specific format. The generic DSO machinery 832 treats the DSO-DATA as an opaque "blob" without attempting to 833 interpret it. Interpretation of the meaning of the DSO-DATA for a 834 particular DSO-TYPE is the responsibility of the software that 835 implements that DSO-TYPE. 837 5.2.2.2. Request TLVs 839 The first TLV in a DSO request message or unacknowledged message is 840 the "Primary TLV" and indicates the operation to be performed. A DSO 841 request message or unacknowledged message MUST contain at at least 842 one TLV, the Primary TLV. 844 Immediately following the Primary TLV, a DSO request message or 845 unacknowledged message MAY contain one or more "Additional TLVs", 846 which specify additional parameters relating to the operation. 848 5.2.2.3. Response TLVs 850 Depending on the operation, a DSO response message MAY contain no 851 TLVs, because it is simply a response to a previous request message, 852 and the MESSAGE ID in the header is sufficient to identify the 853 request in question. Or it may contain a single response TLV, with 854 the same DSO-TYPE as the Primary TLV in the request message. 855 Alternatively it may contain one or more TLVs of other types, or a 856 combination of the above, as appropriate for the information that 857 needs to be communicated. The specification for each DSO TLV 858 determines what TLVs are required in a response to a request using 859 that TLV. 861 If a DSO response is received for an operation where the 862 specification requires that the response carry a particular TLV or 863 TLVs, and the required TLV(s) are not present, then this is a fatal 864 error and the recipient of the defective response message MUST 865 forcibly abort the connection immediately. 867 5.2.2.4. Unrecognized TLVs 869 If DSO request message is received containing an unrecognized Primary 870 TLV, with a nonzero MESSAGE ID (indicating that a response is 871 expected), then the receiver MUST send an error response with 872 matching MESSAGE ID, and RCODE DSOTYPENI ([TBA2] tentatively 11). 873 The error response MUST NOT contain a copy of the unrecognized 874 Primary TLV. 876 If DSO unacknowledged message is received containing an unrecognized 877 Primary TLV, with a zero MESSAGE ID (indicating that no response is 878 expected), then this is a fatal error and the recipient MUST forcibly 879 abort the connection immediately. 881 If a DSO request message or unacknowledged message is received where 882 the Primary TLV is recognized, containing one or more unrecognized 883 Additional TLVs, the unrecognized Additional TLVs MUST be silently 884 ignored, and the remainder of the message is interpreted and handled 885 as if the unrecognized parts were not present. 887 Similarly, if a DSO response message is received containing one or 888 more unrecognized TLVs, the unrecognized TLVs MUST be silently 889 ignored, and the remainder of the message is interpreted and handled 890 as if the unrecognized parts were not present. 892 5.2.3. EDNS(0) and TSIG 894 Since the ARCOUNT field MUST be zero, a DSO message MUST NOT contain 895 an EDNS(0) option in the additional records section. If 896 functionality provided by current or future EDNS(0) options is 897 desired for DSO messages, one or more new DSO TLVs need to be defined 898 to carry the necessary information. 900 For example, the EDNS(0) Padding Option [RFC7830] used for security 901 purposes is not permitted in a DSO message, so if message padding is 902 desired for DSO messages then the Encryption Padding TLV described in 903 Section 7.3 MUST be used. 905 Similarly, a DSO message MUST NOT contain a TSIG record. A TSIG 906 record in a conventional DNS message is added as the last record in 907 the additional records section, and carries a signature computed over 908 the preceding message content. Since DSO data appears *after* the 909 additional records section, it would not be included in the signature 910 calculation. If use of signatures with DSO messages becomes 911 necessary in the future, a new DSO TLV needs to be defined to perform 912 this function. 914 Note however that, while DSO *messages* cannot include EDNS(0) or 915 TSIG records, a DSO *session* is typically used to carry a whole 916 series of DNS messages of different kinds, including DSO messages, 917 and other DNS message types like Query [RFC1034] [RFC1035] and Update 918 [RFC2136], and those messages can carry EDNS(0) and TSIG records. 920 Although messages may contain other EDNS(0) options as appropriate, 921 this specification explicitly prohibits use of the EDNS(0) TCP 922 Keepalive Option [RFC7828] in *any* messages sent on a DSO Session 923 (because it is obsoleted by the functionality provided by the DSO 924 Keepalive operation). If any message sent on a DSO Session contains 925 an EDNS(0) TCP Keepalive Option this is a fatal error and the 926 recipient of the defective message MUST forcibly abort the connection 927 immediately. 929 5.3. Message Handling 931 The initiator MUST set the value of the QR bit in the DNS header to 932 zero (0), and the responder MUST set it to one (1). 934 As described above in Section 5.2.1 whether an outgoing message with 935 QR=0 is unacknowledged or acknowledged is determined by the 936 specification for the Primary TLV, which in turn determines whether 937 the MESSAGE ID field in that outgoing message will be zero or 938 nonzero. 940 A DSO unacknowledged message has both the QR bit and the MESSAGE ID 941 field set to zero, and MUST NOT elicit a response. 943 Every DSO request message (QR=0) with a nonzero MESSAGE ID field is 944 an acknowledged DSO request, and MUST elicit a corresponding response 945 (QR=1), which MUST have the same MESSAGE ID in the DNS message header 946 as in the corresponding request. 948 Valid DSO request messages sent by the client with a nonzero MESSAGE 949 ID field elicit a response from the server, and Valid DSO request 950 messages sent by the server with a nonzero MESSAGE ID field elicit a 951 response from the client. 953 The namespaces of 16-bit MESSAGE IDs are independent in each 954 direction. This means it is *not* an error for both client and 955 server to send request messages at the same time as each other, using 956 the same MESSAGE ID, in different directions. This simplification is 957 necessary in order for the protocol to be implementable. It would be 958 infeasible to require the client and server to coordinate with each 959 other regarding allocation of new unique MESSAGE IDs. It is also not 960 necessary to require the client and server to coordinate with each 961 other regarding allocation of new unique MESSAGE IDs. The value of 962 the 16-bit MESSAGE ID combined with the identity of the initiator 963 (client or server) is sufficient to unambiguously identify the 964 operation in question. This can be thought of as a 17-bit message 965 identifier space, using message identifiers 0x00001-0x0FFFF for 966 client-to-server DSO request messages, and message identifiers 967 0x10001-0x1FFFF for server-to-client DSO request messages. The 968 least-significant 16 bits are stored explicitly in the MESSAGE ID 969 field of the DSO message, and the most-significant bit is implicit 970 from the direction of the message. 972 As described above in Section 5.2.1, an initiator MUST NOT reuse a 973 MESSAGE ID that it already has in use for an outstanding request 974 (unless specified otherwise by the relevant specification for the 975 DSO-TYPE in question). At the very least, this means that a MESSAGE 976 ID MUST NOT be reused in a particular direction on a particular DSO 977 Session while the initiator is waiting for a response to a previous 978 request using that MESSAGE ID on that DSO Session (unless specified 979 otherwise by the relevant specification for the DSO-TYPE in 980 question), and for a long-lived operation the MESSAGE ID for the 981 operation MUST NOT be reused while that operation remains active. 983 If a client or server receives a response (QR=1) where the MESSAGE ID 984 is zero, or is any other value that does not match the MESSAGE ID of 985 any of its outstanding operations, this is a fatal error and the 986 recipient MUST forcibly abort the connection immediately. 988 5.3.1. Error Responses 990 When a DSO unacknowledged message is unsuccessful for some reason, 991 the responder immediately aborts the connection. 993 When a DSO request message is unsuccessful for some reason, the 994 responder returns an error code to the initiator. 996 In the case of a server returning an error code to a client in 997 response to an unsuccessful DSO request message, the server MAY 998 choose to end the DSO Session, or MAY choose to allow the DSO Session 999 to remain open. For error conditions that only affect the single 1000 operation in question, the server SHOULD return an error response to 1001 the client and leave the DSO Session open for further operations. 1003 For error conditions that are likely to make all operations 1004 unsuccessful in the immediate future, the server SHOULD return an 1005 error response to the client and then end the DSO Session by sending 1006 a Retry Delay message, as described in Section 6.6.1. 1008 Upon receiving an error response from the server, a client SHOULD NOT 1009 automatically close the DSO Session. An error relating to one 1010 particular operation on a DSO Session does not necessarily imply that 1011 all other operations on that DSO Session have also failed, or that 1012 future operations will fail. The client should assume that the 1013 server will make its own decision about whether or not to end the DSO 1014 Session, based on the server's determination of whether the error 1015 condition pertains to this particular operation, or would also apply 1016 to any subsequent operations. If the server does not end the DSO 1017 Session by sending the client a Retry Delay message (Section 6.6.1) 1018 then the client SHOULD continue to use that DSO Session for 1019 subsequent operations. 1021 5.4. DSO Response Generation 1023 With most TCP implementations, for DSO requests that generate a 1024 response, the TCP data acknowledgement (generated because data has 1025 been received by TCP), the TCP window update (generated because TCP 1026 has delivered that data to the receiving software), and the DSO 1027 response (generated by the receiving application-layer software 1028 itself) are all combined into a single IP packet. Combining these 1029 three elements into a single IP packet can give a significant 1030 improvement in network efficiency. 1032 For DSO requests that do not generate a response, the TCP 1033 implementation generally doesn't have any way to know that no 1034 response will be forthcoming, so it waits fruitlessly for the 1035 application-layer software to generate a response, until the Delayed 1036 ACK timer fires [RFC1122] (typically 200 milliseconds) and only then 1037 does it send the TCP ACK and window update. In conjunction with 1038 Nagle's Algorithm at the sender, this can delay the sender's 1039 transmission of its next (non-full-sized) TCP segment, while the 1040 sender is waiting for its previous (non-full-sized) TCP segment to be 1041 acknowledged, which won't happen until the Delayed ACK timer fires. 1042 Nagle's Algorithm exists to combine multiple small application writes 1043 into more-efficient large TCP segments, to guard against wasteful use 1044 of the network by applications that would otherwise transmit a stream 1045 of small TCP segments, but in this case Nagle's Algorithm (created to 1046 improve network efficiency) can interact badly with TCP's Delayed ACK 1047 feature (also created to improve network efficiency) [NagleDA] with 1048 the result of delaying some messages by up to 200 milliseconds. 1050 Possible mitigations for this problem include: 1052 o Disable Nagle's Algorithm at the sender. This is not great, 1053 because it results in less efficient use of the network. 1055 o Disable Delayed ACK at the receiver. This is not great, 1056 because it results in less efficient use of the network. 1058 o Adding padding data to fill the segment. This is not great, 1059 because it uses additional bandwidth. 1061 o Use a networking API that lets the receiver signal to the TCP 1062 implementation that the receiver has received and processed a 1063 client request for which it will not be generating any immediate 1064 response. This allows the TCP implementation to operate 1065 efficiently in both cases; for requests that generate a response, 1066 the TCP ACK, window update, and DSO response are transmitted 1067 together in a single TCP segment, and for requests that do not 1068 generate a response, the application-layer software informs the 1069 TCP implementation that it should go ahead and send the TCP ACK 1070 and window update immediately, without waiting for the Delayed ACK 1071 timer. Unfortunately it is not known at this time which (if any) 1072 of the widely-available networking APIs currently include this 1073 capability. 1075 5.5. Responder-Initiated Operation Cancellation 1077 This document, the base specification for DNS Stateful Operations, 1078 does not itself define any long-lived operations, but it defines a 1079 framework for supporting long-lived operations, such as Push 1080 Notification subscriptions [I-D.ietf-dnssd-push] and Discovery Relay 1081 interface subscriptions [I-D.ietf-dnssd-mdns-relay]. 1083 Generally speaking, a long-lived operation is initiated by the 1084 initiator, and, if successful, remains active until the initiator 1085 terminates the operation. 1087 However, it is possible that a long-lived operation may be valid at 1088 the time it was initiated, but then a later change of circumstances 1089 may render that previously valid operation invalid. 1091 For example, a long-lived client operation may pertain to a name that 1092 the server is authoritative for, but then the server configuration is 1093 changed such that it is no longer authoritative for that name. 1095 In such cases, instead of terminating the entire session it may be 1096 desirable for the responder to be able to cancel selectively only 1097 those operations that have become invalid. 1099 The responder performs this selective cancellation by sending a new 1100 response message, with the MESSAGE ID field containing the MESSAGE ID 1101 of the long-lived operation that is to be terminated (that it had 1102 previously acknowledged with a NOERROR RCODE), and the RCODE field of 1103 the new response message giving the reason for cancellation. 1105 After a response message with nonzero RCODE has been sent, that 1106 operation has been terminated from the responder's point of view, and 1107 the responder sends no more messages relating to that operation. 1109 After a response message with nonzero RCODE has been received by the 1110 initiator, that operation has been terminated from the initiator's 1111 point of view, and the cancelled operation's MESSAGE ID is now free 1112 for reuse. 1114 6. DSO Session Lifecycle and Timers 1116 6.1. DSO Session Initiation 1118 A DSO Session begins as described in Section 5.1. 1120 The client may perform as many DNS operations as it wishes using the 1121 newly created DSO Session. Operations SHOULD be pipelined (i.e., the 1122 client doesn't need wait for a response before sending the next 1123 message). The server MUST act on messages in the order they are 1124 transmitted, but responses to those messages SHOULD be sent out of 1125 order when appropriate. 1127 6.2. DSO Session Timeouts 1129 Two timeout values are associated with a DSO Session: the inactivity 1130 timeout, and the keepalive interval. Both values are communicated in 1131 the same TLV, the DSO Keepalive TLV (Section 7.1). 1133 The first timeout value, the inactivity timeout, is the maximum time 1134 for which a client may speculatively keep a DSO Session open in the 1135 expectation that it may have future requests to send to that server. 1137 The second timeout value, the keepalive interval, is the maximum 1138 permitted interval between messages if the client wishes to keep the 1139 DSO Session alive. 1141 The two timeout values are independent. The inactivity timeout may 1142 be lower, the same, or higher than the keepalive interval, though in 1143 most cases the inactivity timeout is expected to be shorter than the 1144 keepalive interval. 1146 A shorter inactivity timeout with a longer keepalive interval signals 1147 to the client that it should not speculatively keep an inactive DSO 1148 Session open for very long without reason, but when it does have an 1149 active reason to keep a DSO Session open, it doesn't need to be 1150 sending an aggressive level of keepalive traffic to maintain that 1151 session. 1153 A longer inactivity timeout with a shorter keepalive interval signals 1154 to the client that it may speculatively keep an inactive DSO Session 1155 open for a long time, but to maintain that inactive DSO Session it 1156 should be sending a lot of keepalive traffic. This configuration is 1157 expected to be less common. 1159 In the usual case where the inactivity timeout is shorter than the 1160 keepalive interval, it is only when a client has a very long-lived, 1161 low-traffic, operation that the keepalive interval comes into play, 1162 to ensure that a sufficient residual amount of traffic is generated 1163 to maintain NAT and firewall state and to assure client and server 1164 that they still have connectivity to each other. 1166 On a new DSO Session, if no explicit DSO Keepalive message exchange 1167 has taken place, the default value for both timeouts is 15 seconds. 1169 For both timeouts, lower values of the timeout result in higher 1170 network traffic and higher CPU load on the server. 1172 6.3. Inactive DSO Sessions 1174 At both servers and clients, the generation or reception of any 1175 complete DNS message, including DNS requests, responses, updates, or 1176 DSO messages, resets both timers for that DSO Session, with the 1177 exception that a DSO Keepalive message resets only the keepalive 1178 timer, not the inactivity timeout timer. 1180 In addition, for as long as the client has an outstanding operation 1181 in progress, the inactivity timer remains cleared, and an inactivity 1182 timeout cannot occur. 1184 For short-lived DNS operations like traditional queries and updates, 1185 an operation is considered in progress for the time between request 1186 and response, typically a period of a few hundred milliseconds at 1187 most. At the client, the inactivity timer is cleared upon 1188 transmission of a request and remains cleared until reception of the 1189 corresponding response. At the server, the inactivity timer is 1190 cleared upon reception of a request and remains cleared until 1191 transmission of the corresponding response. 1193 For long-lived DNS Stateful operations (such as a Push Notification 1194 subscription [I-D.ietf-dnssd-push] or a Discovery Relay interface 1195 subscription [I-D.ietf-dnssd-mdns-relay]), an operation is considered 1196 in progress for as long as the operation is active, until it is 1197 cancelled. This means that a DSO Session can exist, with active 1198 operations, with no messages flowing in either direction, for far 1199 longer than the inactivity timeout, and this is not an error. This 1200 is why there are two separate timers: the inactivity timeout, and the 1201 keepalive interval. Just because a DSO Session has no traffic for an 1202 extended period of time does not automatically make that DSO Session 1203 "inactive", if it has an active operation that is awaiting events. 1205 6.4. The Inactivity Timeout 1207 The purpose of the inactivity timeout is for the server to balance 1208 its trade off between the costs of setting up new DSO Sessions and 1209 the costs of maintaining inactive DSO Sessions. A server with 1210 abundant DSO Session capacity can offer a high inactivity timeout, to 1211 permit clients to keep a speculative DSO Session open for a long 1212 time, to save the cost of establishing a new DSO Session for future 1213 communications with that server. A server with scarce memory 1214 resources can offer a low inactivity timeout, to cause clients to 1215 promptly close DSO Sessions whenever they have no outstanding 1216 operations with that server, and then create a new DSO Session later 1217 when needed. 1219 6.4.1. Closing Inactive DSO Sessions 1221 When a connection's inactivity timeout is reached the client MUST 1222 begin closing the idle connection, but a client is not required to 1223 keep an idle connection open until the inactivity timeout is reached. 1224 A client MAY close a DSO Session at any time, at the client's 1225 discretion. If a client determines that it has no current or 1226 reasonably anticipated future need for a currently inactive DSO 1227 Session, then the client SHOULD gracefully close that connection. 1229 If, at any time during the life of the DSO Session, the inactivity 1230 timeout value (i.e., 15 seconds by default) elapses without there 1231 being any operation active on the DSO Session, the client MUST close 1232 the connection gracefully. 1234 If, at any time during the life of the DSO Session, twice the 1235 inactivity timeout value (i.e., 30 seconds by default), or five 1236 seconds, if twice the inactivity timeout value is less than five 1237 seconds, elapses without there being any operation active on the DSO 1238 Session, the server SHOULD consider the client delinquent, and SHOULD 1239 forcibly abort the DSO Session. 1241 In this context, an operation being active on a DSO Session includes 1242 a query waiting for a response, an update waiting for a response, or 1243 an active long-lived operation, but not a DSO Keepalive message 1244 exchange itself. A DSO Keepalive message exchange resets only the 1245 keepalive interval timer, not the inactivity timeout timer. 1247 If the client wishes to keep an inactive DSO Session open for longer 1248 than the default duration then it uses the DSO Keepalive message to 1249 request longer timeout values, as described in Section 7.1. 1251 6.4.2. Values for the Inactivity Timeout 1253 For the inactivity timeout value, lower values result in more 1254 frequent DSO Session teardown and re-establishment. Higher values 1255 result in lower traffic and lower CPU load on the server, but higher 1256 memory burden to maintain state for inactive DSO Sessions. 1258 A server may dictate any value it chooses for the inactivity timeout 1259 (either in a response to a client-initiated request, or in a server- 1260 initiated message) including values under one second, or even zero. 1262 An inactivity timeout of zero informs the client that it should not 1263 speculatively maintain idle connections at all, and as soon as the 1264 client has completed the operation or operations relating to this 1265 server, the client should immediately begin closing this session. 1267 A server will abort an idle client session after twice the inactivity 1268 timeout value, or five seconds, whichever is greater. In the case of 1269 a zero inactivity timeout value, this means that if a client fails to 1270 close an idle client session then the server will forcibly abort the 1271 idle session after five seconds. 1273 An inactivity timeout of 0xFFFFFFFF represents "infinity" and informs 1274 the client that it may keep an idle connection open as long as it 1275 wishes. Note that after granting an unlimited inactivity timeout in 1276 this way, at any point the server may revise that inactivity timeout 1277 by sending a new Keepalive message dictating new Session Timeout 1278 values to the client. 1280 The largest *finite* inactivity timeout supported by the current DSO 1281 Keepalive TLV is 0xFFFFFFFE (2^32-2 milliseconds, approximately 49.7 1282 days). 1284 6.5. The Keepalive Interval 1286 The purpose of the keepalive interval is to manage the generation of 1287 sufficient messages to maintain state in middleboxes (such at NAT 1288 gateways or firewalls) and for the client and server to periodically 1289 verify that they still have connectivity to each other. This allows 1290 them to clean up state when connectivity is lost, and to establish a 1291 new session if appropriate. 1293 6.5.1. Keepalive Interval Expiry 1295 If, at any time during the life of the DSO Session, the keepalive 1296 interval value (i.e., 15 seconds by default) elapses without any DNS 1297 messages being sent or received on a DSO Session, the client MUST 1298 take action to keep the DSO Session alive, by sending a DSO Keepalive 1299 message (Section 7.1). A DSO Keepalive message exchange resets only 1300 the keepalive timer, not the inactivity timer. 1302 If a client disconnects from the network abruptly, without cleanly 1303 closing its DSO Session, perhaps leaving a long-lived operation 1304 uncancelled, the server learns of this after failing to receive the 1305 required keepalive traffic from that client. If, at any time during 1306 the life of the DSO Session, twice the keepalive interval value 1307 (i.e., 30 seconds by default) elapses without any DNS messages being 1308 sent or received on a DSO Session, the server SHOULD consider the 1309 client delinquent, and SHOULD forcibly abort the DSO Session. 1311 6.5.2. Values for the Keepalive Interval 1313 For the keepalive interval value, lower values result in a higher 1314 volume of keepalive traffic. Higher values of the keepalive interval 1315 reduce traffic and CPU load, but have minimal effect on the memory 1316 burden at the server, because clients keep a DSO Session open for the 1317 same length of time (determined by the inactivity timeout) regardless 1318 of the level of keepalive traffic required. 1320 It may be appropriate for clients and servers to select different 1321 keepalive interval values depending on the nature of the network they 1322 are on. 1324 A corporate DNS server that knows it is serving only clients on the 1325 internal network, with no intervening NAT gateways or firewalls, can 1326 impose a higher keepalive interval, because frequent keepalive 1327 traffic is not required. 1329 A public DNS server that is serving primarily residential consumer 1330 clients, where it is likely there will be a NAT gateway on the path, 1331 may impose a lower keepalive interval, to generate more frequent 1332 keepalive traffic. 1334 A smart client may be adaptive to its environment. A client using a 1335 private IPv4 address [RFC1918] to communicate with a DNS server at an 1336 address outside that IPv4 private address block, may conclude that 1337 there is likely to be a NAT gateway on the path, and accordingly 1338 request a lower keepalive interval. 1340 By default it is RECOMMENDED that clients request, and servers grant, 1341 a keepalive interval of 60 minutes. This keepalive interval provides 1342 for reasonably timely detection if a client abruptly disconnects 1343 without cleanly closing the session, and is sufficient to maintain 1344 state in firewalls and NAT gateways that follow the IETF recommended 1345 Best Current Practice that the "established connection idle-timeout" 1346 used by middleboxes be at least 2 hours 4 minutes [RFC5382]. 1348 Note that the lower the keepalive interval value, the higher the load 1349 on client and server. For example, a hypothetical keepalive interval 1350 value of 100ms would result in a continuous stream of at least ten 1351 messages per second, in both directions, to keep the DSO Session 1352 alive. And, in this extreme example, a single packet loss and 1353 retransmission over a long path could introduce a momentary pause in 1354 the stream of messages, long enough to cause the server to 1355 overzealously abort the connection. 1357 Because of this concern, the server MUST NOT send a Keepalive message 1358 (either a response to a client-initiated request, or a server- 1359 initiated message) with a keepalive interval value less than ten 1360 seconds. If a client receives a Keepalive message specifying a 1361 keepalive interval value less than ten seconds this is a fatal error 1362 and the client MUST forcibly abort the connection immediately. 1364 A keepalive interval value of 0xFFFFFFFF represents "infinity" and 1365 informs the client that it should generate no keepalive traffic. 1366 Note that after signaling that the client should generate no 1367 keepalive traffic in this way, at any point the server may revise 1368 that keepalive traffic requirement by sending a new Keepalive message 1369 dictating new Session Timeout values to the client. 1371 The largest *finite* keepalive interval supported by the current DSO 1372 Keepalive TLV is 0xFFFFFFFE (2^32-2 milliseconds, approximately 49.7 1373 days). 1375 6.6. Server-Initiated Session Termination 1377 In addition to cancelling individual long-lived operations 1378 selectively (Section 5.5) there are also occasions where a server may 1379 need to terminate one or more entire sessions. An entire session may 1380 need to be terminated if the client is defective in some way, or 1381 departs from the network without closing its session. Sessions may 1382 also need to be terminated if the server becomes overloaded, or if 1383 the server is reconfigured and lacks the ability to be selective 1384 about which operations need to be cancelled. 1386 This section discusses various reasons a session may be terminated, 1387 and the mechanisms for doing so. 1389 Normally a server MUST NOT close a DSO Session with a client. A 1390 server only causes a DSO Session to be ended in the exceptional 1391 circumstances outlined below. In normal operation, closing a DSO 1392 Session is the client's responsibility. The client makes the 1393 determination of when to close a DSO Session based on an evaluation 1394 of both its own needs, and the inactivity timeout value dictated by 1395 the server. 1397 Some of the exceptional situations in which a server may terminate a 1398 DSO Session include: 1400 o The server application software or underlying operating system is 1401 shutting down or restarting. 1403 o The server application software terminates unexpectedly (perhaps 1404 due to a bug that makes it crash). 1406 o The server is undergoing a reconfiguration or maintenance 1407 procedure, that, due to the way the server software is 1408 implemented, requires clients to be disconnected. For example, 1409 some software is implemented such that it reads a configuration 1410 file at startup, and changing the server's configuration entails 1411 modifying the configuration file and then killing and restarting 1412 the server software, which generally entails a loss of network 1413 connections. 1415 o The client fails to meets its obligation to generate the required 1416 keepalive traffic, or to close an inactive session by the 1417 prescribed time (twice the time interval dictated by the server, 1418 or five seconds, whichever is greater, as described in 1419 Section 6.2). 1421 o The client sends a grossly invalid or malformed request that is 1422 indicative of a seriously defective client implementation. 1424 o The server is over capacity and needs to shed some load. 1426 6.6.1. Server-Initiated Retry Delay Message 1428 In the cases described above where a server elects to terminate a DSO 1429 Session, it could do so simply by forcibly aborting the connection. 1430 However, if it did this the likely behavior of the client might be 1431 simply to to treat this as a network failure and reconnect 1432 immediately, putting more burden on the server. 1434 Therefore, to avoid this reconnection implosion, a server SHOULD 1435 instead choose to shed client load by sending a Retry Delay message, 1436 with an appropriate RCODE value informing the client of the reason 1437 the DSO Session needs to be terminated. The format of the Retry 1438 Delay TLV, and the interpretations of the various RCODE values, are 1439 described in Section 7.2. After sending a Retry Delay message, the 1440 server MUST NOT send any further messages on that DSO Session. 1442 Upon receipt of a Retry Delay message from the server, the client 1443 MUST make note of the reconnect delay for this server, and then 1444 immediately close the connection gracefully. 1446 After sending a Retry Delay message the server SHOULD allow the 1447 client five seconds to close the connection, and if the client has 1448 not closed the connection after five seconds then the server SHOULD 1449 forcibly abort the connection. 1451 A Retry Delay message MUST NOT be initiated by a client. If a server 1452 receives a Retry Delay message this is a fatal error and the server 1453 MUST forcibly abort the connection immediately. 1455 6.6.1.1. Outstanding Operations 1457 At the instant a server chooses to initiate a Retry Delay message 1458 there may be DNS requests already in flight from client to server on 1459 this DSO Session, which will arrive at the server after its Retry 1460 Delay message has been sent. The server MUST silently ignore such 1461 incoming requests, and MUST NOT generate any response messages for 1462 them. When the Retry Delay message from the server arrives at the 1463 client, the client will determine that any DNS requests it previously 1464 sent on this DSO Session, that have not yet received a response, now 1465 will certainly not be receiving any response. Such requests should 1466 be considered failed, and should be retried at a later time, as 1467 appropriate. 1469 In the case where some, but not all, of the existing operations on a 1470 DSO Session have become invalid (perhaps because the server has been 1471 reconfigured and is no longer authoritative for some of the names), 1472 but the server is terminating all affected DSO Sessions en masse by 1473 sending them all a Retry Delay message, the RECONNECT DELAY MAY be 1474 zero, indicating that the clients SHOULD immediately attempt to re- 1475 establish operations. 1477 It is likely that some of the attempts will be successful and some 1478 will not, depending on the nature of the reconfiguration. 1480 In the case where a server is terminating a large number of DSO 1481 Sessions at once (e.g., if the system is restarting) and the server 1482 doesn't want to be inundated with a flood of simultaneous retries, it 1483 SHOULD send different RECONNECT delay values to each client. These 1484 adjustments MAY be selected randomly, pseudorandomly, or 1485 deterministically (e.g., incrementing the time value by one tenth of 1486 a second for each successive client, yielding a post-restart 1487 reconnection rate of ten clients per second). 1489 6.6.1.2. Client Reconnection 1491 After a DSO Session is ended by the server (either by sending the 1492 client a Retry Delay message, or by forcibly aborting the underlying 1493 transport connection) the client SHOULD try to reconnect, to that 1494 service instance, or to another suitable service instance, if more 1495 than one is available. If reconnecting to the same service instance, 1496 the client MUST respect the indicated delay, if available, before 1497 attempting to reconnect. 1499 If the service instance will only be out of service for a short 1500 maintenance period, it should use a value a little longer that the 1501 expected maintenance window. It should not default to a very large 1502 delay value, or clients may not attempt to reconnect after it resumes 1503 service. 1505 If a particular service instance does not want a client to reconnect 1506 ever (perhaps the service instance is being de-commissioned), it 1507 SHOULD set the retry delay to the maximum value 0xFFFFFFFF (2^32-1 1508 milliseconds, approximately 49.7 days). It is not possible to 1509 instruct a client to stay away for longer than 49.7 days. If, after 1510 49.7 days, the DNS or other configuration information still indicates 1511 that this is the valid service instance for a particular service, 1512 then clients MAY attempt to reconnect. In reality, if a client is 1513 rebooted or otherwise lose state, it may well attempt to reconnect 1514 before 49.7 days elapses, for as long as the DNS or other 1515 configuration information continues to indicate that this is the 1516 service instance the client should use. 1518 7. Base TLVs for DNS Stateful Operations 1520 This section describes the three base TLVs for DNS Stateful 1521 Operations: Keepalive, Retry Delay, and Encryption Padding. 1523 7.1. Keepalive TLV 1525 The Keepalive TLV (DSO-TYPE=1) performs two functions: to reset the 1526 keepalive timer for the DSO Session, and to establish the values for 1527 the Session Timeouts. 1529 The DSO-DATA for the the Keepalive TLV is as follows: 1531 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 1532 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1533 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1534 | INACTIVITY TIMEOUT (32 bits) | 1535 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1536 | KEEPALIVE INTERVAL (32 bits) | 1537 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1539 INACTIVITY TIMEOUT: The inactivity timeout for the current DSO 1540 Session, specified as a 32-bit unsigned integer, in network (big 1541 endian) byte order, in units of milliseconds. This is the timeout 1542 at which the client MUST begin closing an inactive DSO Session. 1543 The inactivity timeout can be any value of the server's choosing. 1544 If the client does not gracefully close an inactive DSO Session, 1545 then after twice this interval, or five seconds, whichever is 1546 greater, the server will forcibly abort the connection. 1548 KEEPALIVE INTERVAL: The keepalive interval for the current DSO 1549 Session, specified as a 32-bit unsigned integer, in network (big 1550 endian) byte order, in units of milliseconds. This is the 1551 interval at which a client MUST generate keepalive traffic to 1552 maintain connection state. The keepalive interval MUST NOT be 1553 less than ten seconds. If the client does not generate the 1554 mandated keepalive traffic, then after twice this interval the 1555 server will forcibly abort the connection. Since the minimum 1556 allowed keepalive interval is ten seconds, the minimum time at 1557 which a server will forcibly disconnect a client for failing to 1558 generate the mandated keepalive traffic is twenty seconds. 1560 The transmission or reception of DSO Keepalive messages (i.e., 1561 messages where the Keepalive TLV is the first TLV) reset only the 1562 keepalive timer, not the inactivity timer. The reason for this is 1563 that periodic Keepalive messages are sent for the sole purpose of 1564 keeping a DSO Session alive, when that DSO Session has current or 1565 recent non-maintenance activity that warrants keeping that DSO 1566 Session alive. Sending keepalive traffic itself is not considered a 1567 client activity; it is considered a maintenance activity that is 1568 performed in service of other client activities. If keepalive 1569 traffic itself were to reset the inactivity timer, then that would 1570 create a circular livelock where keepalive traffic would be sent 1571 indefinitely to keep a DSO Session alive, where the only activity on 1572 that DSO Session would be the keepalive traffic keeping the DSO 1573 Session alive so that further keepalive traffic can be sent. For a 1574 DSO Session to be considered active, it must be carrying something 1575 more than just keepalive traffic. This is why merely sending or 1576 receiving a Keepalive message does not reset the inactivity timer. 1578 When sent by a client, the Keepalive request message MUST be sent as 1579 an acknowledged request, with a nonzero MESSAGE ID. If a server 1580 receives a Keepalive DSO message with a zero MESSAGE ID then this is 1581 a fatal error and the server MUST forcibly abort the connection 1582 immediately. The Keepalive request message resets a DSO Session's 1583 keepalive timer, and at the same time communicates to the server the 1584 the client's requested Session Timeout values. In a server response 1585 to a client-initiated Keepalive request message, the Session Timeouts 1586 contain the server's chosen values from this point forward in the DSO 1587 Session, which the client MUST respect. This is modeled after the 1588 DHCP protocol, where the client requests a certain lease lifetime 1589 using DHCP option 51 [RFC2132], but the server is the ultimate 1590 authority for deciding what lease lifetime is actually granted. 1592 When a client is sending its second and subsequent Keepalive DSO 1593 requests to the server, the client SHOULD continue to request its 1594 preferred values each time. This allows flexibility, so that if 1595 conditions change during the lifetime of a DSO Session, the server 1596 can adapt its responses to better fit the client's needs. 1598 Once a DSO Session is in progress (Section 5.1) a Keepalive message 1599 MAY be initiated by a server. When sent by a server, the Keepalive 1600 message MUST be sent as an unacknowledged message, with the MESSAGE 1601 ID set to zero. The client MUST NOT generate a response to a server- 1602 initiated DSO Keepalive message. If a client receives a Keepalive 1603 request message with a nonzero MESSAGE ID then this is a fatal error 1604 and the client MUST forcibly abort the connection immediately. The 1605 Keepalive unacknowledged message from the server resets a DSO 1606 Session's keepalive timer, and at the same time unilaterally informs 1607 the client of the new Session Timeout values to use from this point 1608 forward in this DSO Session. No client DSO response message to this 1609 unilateral declaration is required or allowed. 1611 The Keepalive TLV is not used as an Additional TLV. 1613 In response messages the Keepalive TLV is used only as a Response 1614 Primary TLV, replying to a Keepalive request message from the client. 1615 A Keepalive TLV MUST NOT be added as to other responses a Response 1616 Additional TLV. If the server wishes to update a client's Session 1617 Timeout values other than in response to a Keepalive request message 1618 from the client, then it does so by sending an unacknowledged 1619 Keepalive message of its own, as described above. 1621 It is not required that the Keepalive TLV be used in every DSO 1622 Session. While many DNS Stateful operations will be used in 1623 conjunction with a long-lived session state, not all DNS Stateful 1624 operations require long-lived session state, and in some cases the 1625 default 15-second value for both the inactivity timeout and keepalive 1626 interval may be perfectly appropriate. However, note that for 1627 clients that implement only the DSO-TYPEs defined in this document, a 1628 Keepalive request message is the only way for a client to initiate a 1629 DSO Session. 1631 7.1.1. Client handling of received Session Timeout values 1633 When a client receives a response to its client-initiated DSO 1634 Keepalive message, or receives a server-initiated DSO Keepalive 1635 message, the client has then received Session Timeout values dictated 1636 by the server. The two timeout values contained in the DSO Keepalive 1637 TLV from the server may each be higher, lower, or the same as the 1638 respective Session Timeout values the client previously had for this 1639 DSO Session. 1641 In the case of the keepalive timer, the handling of the received 1642 value is straightforward. The act of receiving the message 1643 containing the DSO Keepalive TLV itself resets the keepalive timer 1644 and updates the keepalive interval for the DSO Session. The new 1645 keepalive interval indicates the maximum time that may elapse before 1646 another message must be sent or received on this DSO Session, if the 1647 DSO Session is to remain alive. 1649 In the case of the inactivity timeout, the handling of the received 1650 value is a little more subtle, though the meaning of the inactivity 1651 timeout remains as specified -- it still indicates the maximum 1652 permissible time allowed without useful activity on a DSO Session. 1653 The act of receiving the message containing the DSO Keepalive TLV 1654 does not itself reset the inactivity timer. The time elapsed since 1655 the last useful activity on this DSO Session is unaffected by 1656 exchange of DSO Keepalive messages. The new inactivity timeout value 1657 in the DSO Keepalive TLV in the received message does update the 1658 timeout associated with the running inactivity timer; that becomes 1659 the new maximum permissible time without activity on a DSO Session. 1661 o If the current inactivity timer value is less than the new 1662 inactivity timeout, then the DSO Session may remain open for now. 1663 When the inactivity timer value reaches the new inactivity 1664 timeout, the client MUST then begin closing the DSO Session, as 1665 described above. 1667 o If the current inactivity timer value is equal to the new 1668 inactivity timeout, then this DSO Session has been inactive for 1669 exactly as long as the server will permit, and now the client MUST 1670 immediately begin closing this DSO Session. 1672 o If the current inactivity timer value is already greater than the 1673 new inactivity timeout, then this DSO Session has already been 1674 inactive for longer than the server permits, and the client MUST 1675 immediately begin closing this DSO Session. 1677 o If the current inactivity timer value is already more than twice 1678 the new inactivity timeout, then the client is immediately 1679 considered delinquent (this DSO Session is immediately eligible to 1680 be forcibly terminated by the server) and the client MUST 1681 immediately begin closing this DSO Session. However if a server 1682 abruptly reduces the inactivity timeout in this way, then, to give 1683 the client time to close the connection gracefully before the 1684 server resorts to forcibly aborting it, the server SHOULD give the 1685 client an additional grace period of one quarter of the new 1686 inactivity timeout, or five seconds, whichever is greater. 1688 7.1.2. Relation to EDNS(0) TCP Keepalive Option 1690 The inactivity timeout value in the Keepalive TLV (DSO-TYPE=1) has 1691 similar intent to the EDNS(0) TCP Keepalive Option [RFC7828]. A 1692 client/server pair that supports DSO MUST NOT use the EDNS(0) TCP 1693 KeepAlive option within any message after a DSO Session has been 1694 established. Once a DSO Session has been established, if either 1695 client or server receives a DNS message over the DSO Session that 1696 contains an EDNS(0) TCP Keepalive option, this is a fatal error and 1697 the receiver of the EDNS(0) TCP Keepalive option MUST forcibly abort 1698 the connection immediately. 1700 7.2. Retry Delay TLV 1702 The Retry Delay TLV (DSO-TYPE=2) can be used as a Primary TLV 1703 (unacknowledged) in a server-to-client message, or as a Response 1704 Additional TLV in either direction. 1706 The DSO-DATA for the the Retry Delay TLV is as follows: 1708 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 1709 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1710 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1711 | RETRY DELAY (32 bits) | 1712 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1714 RETRY DELAY: A time value, specified as a 32-bit unsigned integer, 1715 in network (big endian) byte order, in units of milliseconds, 1716 within which the initiator MUST NOT retry this operation, or retry 1717 connecting to this server. Recommendations for the RETRY DELAY 1718 value are given in Section 6.6.1. 1720 7.2.1. Retry Delay TLV used as a Primary TLV 1722 When sent from server to client, the Retry Delay TLV is used as the 1723 Primary TLV in an unacknowledged message. It is used by a server to 1724 instruct a client to close the DSO Session and underlying connection, 1725 and not to reconnect for the indicated time interval. 1727 In this case it applies to the DSO Session as a whole, and the client 1728 MUST begin closing the DSO Session, as described in Section 6.6.1. 1729 The RCODE in the message header SHOULD indicate the principal reason 1730 for the termination: 1732 o NOERROR indicates a routine shutdown or restart. 1734 o FORMERR indicates that the client requests are too badly malformed 1735 for the session to continue. 1737 o SERVFAIL indicates that the server is overloaded due to resource 1738 exhaustion and needs to shed load. 1740 o REFUSED indicates that the server has been reconfigured, and at 1741 this time it is now unable to perform one or more of the long- 1742 lived client operations that were previously being performed on 1743 this DSO Session. 1745 o NOTAUTH indicates that the server has been reconfigured and at 1746 this time it is now unable to perform one or more of the long- 1747 lived client operations that were previously being performed on 1748 this DSO Session because it does not have authority over the names 1749 in question (for example, a DNS Push Notification server could be 1750 reconfigured such that is is no longer accepting DNS Push 1751 Notification requests for one or more of the currently subscribed 1752 names). 1754 This document specifies only these RCODE values for Retry Delay 1755 message. Servers sending Retry Delay messages SHOULD use one of 1756 these values. However, future circumstances may create situations 1757 where other RCODE values are appropriate in Retry Delay messages, so 1758 clients MUST be prepared to accept Retry Delay messages with any 1759 RCODE value. 1761 In some cases, when a server sends a Retry Delay message to a client, 1762 there may be more than one reason for the server wanting to end the 1763 session. Possibly the configuration could have been changed such 1764 that some long-lived client operations can no longer be continued due 1765 to policy (REFUSED), and other long-lived client operations can no 1766 longer be performed due to the server no longer being authoritative 1767 for those names (NOTAUTH). In such cases the server MAY use any of 1768 the applicable RCODE values, or RCODE=NOERROR (routine shutdown or 1769 restart). 1771 Note that the selection of RCODE value in a Retry Delay message is 1772 not critical, since the RCODE value is generally used only for 1773 information purposes, such as writing to a log file for future human 1774 analysis regarding the nature of the disconnection. Generally 1775 clients do not modify their behavior depending on the RCODE value. 1776 The RETRY DELAY in the message tells the client how long it should 1777 wait before attempting a new connection to this service instance. 1779 For clients that do in some way modify their behavior depending on 1780 the RCODE value, they should treat unknown RCODE values the same as 1781 RCODE=NOERROR (routine shutdown or restart). 1783 A Retry Delay message from server to client is an unacknowledged 1784 message; the MESSAGE ID MUST be set to zero in the outgoing message 1785 and the client MUST NOT send a response. 1787 A client MUST NOT send a Retry Delay DSO request message or DSO 1788 unacknowledged message to a server. If a server receives a DNS 1789 request message (i.e., QR=0) where the Primary TLV is the Retry Delay 1790 TLV, this is a fatal error and the server MUST forcibly abort the 1791 connection immediately. 1793 7.2.2. Retry Delay TLV used as a Response Additional TLV 1795 In the case of a request that returns a nonzero RCODE value, the 1796 responder MAY append a Retry Delay TLV to the response, indicating 1797 the time interval during which the initiator SHOULD NOT attempt this 1798 operation again. 1800 The indicated time interval during which the initiator SHOULD NOT 1801 retry applies only to the failed operation, not to the DSO Session as 1802 a whole. 1804 7.3. Encryption Padding TLV 1806 The Encryption Padding TLV (DSO-TYPE=3) can only be used as an 1807 Additional or Response Additional TLV. It is only applicable when 1808 the DSO Transport layer uses encryption such as TLS. 1810 The DSO-DATA for the the Padding TLV is optional and is a variable 1811 length field containing non-specified values. A DSO-LENGTH of 0 1812 essentially provides for 4 bytes of padding (the minimum amount). 1814 1 1 1 1 1 1 1815 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 1816 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1817 / / 1818 / VARIABLE NUMBER OF BYTES / 1819 / / 1820 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1822 As specified for the EDNS(0) Padding Option [RFC7830] the PADDING 1823 bytes SHOULD be set to 0x00. Other values MAY be used, for example, 1824 in cases where there is a concern that the padded message could be 1825 subject to compression before encryption. PADDING bytes of any value 1826 MUST be accepted in the messages received. 1828 The Encryption Padding TLV may be included in either a DSO request, 1829 response, or both. As specified for the EDNS(0) Padding Option 1830 [RFC7830] if a request is received with an Encryption Padding TLV, 1831 then the response MUST also include an Encryption Padding TLV. 1833 The length of padding is intentionally not specified in this document 1834 and is a function of current best practices with respect to the type 1835 and length of data in the preceding TLVs 1836 [I-D.ietf-dprive-padding-policy]. 1838 8. Summary Highlights 1840 This section summarizes some noteworthy highlights about various 1841 components of the DSO protocol. 1843 8.1. QR bit and MESSAGE ID 1845 In DSO Request Messages the QR bit is 0 and the MESSAGE ID is 1846 nonzero. 1848 In DSO Response Messages the QR bit is 1 and the MESSAGE ID is 1849 nonzero. 1851 In DSO Unacknowledged Messages the QR bit is 0 and the MESSAGE ID is 1852 zero. 1854 The table below illustrates which combinations are legal and how they 1855 are interpreted: 1857 +--------------------------+------------------------+ 1858 | MESSAGE ID zero | MESSAGE ID nonzero | 1859 +--------+--------------------------+------------------------+ 1860 | QR=0 | Unacknowledged Message | Request Message | 1861 +--------+--------------------------+------------------------+ 1862 | QR=1 | Invalid - Fatal Error | Response Message | 1863 +--------+--------------------------+------------------------+ 1865 8.2. TLV Usage 1867 The table below indicates, for each of the three TLVs defined in this 1868 document, whether they are valid in each of ten different contexts. 1870 The first five contexts are requests or unacknowledged messages from 1871 client to server, and the corresponding responses from server back to 1872 client: 1874 o C-P - Primary TLV, sent in DSO Request message, from client to 1875 server, with nonzero MESSAGE ID indicating that this request MUST 1876 generate response message. 1878 o C-U - Primary TLV, sent in DSO Unacknowledged message, from client 1879 to server, with zero MESSAGE ID indicating that this request MUST 1880 NOT generate response message. 1882 o C-A - Additional TLV, optionally added to request message or 1883 unacknowledged message from client to server. 1885 o CRP - Response Primary TLV, included in response message sent back 1886 to the client (in response to a client "C-P" request with nonzero 1887 MESSAGE ID indicating that a response is required) where the DSO- 1888 TYPE of the Response TLV matches the DSO-TYPE of the Primary TLV 1889 in the request. 1891 o CRA - Response Additional TLV, included in response message sent 1892 back to the client (in response to a client "C-P" request with 1893 nonzero MESSAGE ID indicating that a response is required) where 1894 the DSO-TYPE of the Response TLV does not match the DSO-TYPE of 1895 the Primary TLV in the request. 1897 The second five contexts are their counterparts in the opposite 1898 direction: requests or unacknowledged messages from server to client, 1899 and the corresponding responses from client back to server. 1901 +-------------------------+-------------------------+ 1902 | C-P C-U C-A CRP CRA | S-P S-U S-A SRP SRA | 1903 +------------+-------------------------+-------------------------+ 1904 | KeepAlive | X X | X | 1905 +------------+-------------------------+-------------------------+ 1906 | RetryDelay | X | X | 1907 +------------+-------------------------+-------------------------+ 1908 | Padding | X X | X X | 1909 +------------+-------------------------+-------------------------+ 1911 Note that some of the columns in this table are currently empty. The 1912 table provides a template for future TLV definitions to follow. It 1913 is recommended that definitions of future TLVs include a similar 1914 table summarizing the contexts where the new TLV is valid. 1916 9. IANA Considerations 1918 9.1. DSO OPCODE Registration 1920 The IANA is requested to record the value ([TBA1] tentatively) 6 for 1921 the DSO OPCODE in the DNS OPCODE Registry. DSO stands for DNS 1922 Stateful Operations. 1924 9.2. DSO RCODE Registration 1926 The IANA is requested to record the value ([TBA2] tentatively) 11 for 1927 the DSOTYPENI error code in the DNS RCODE Registry. The DSOTYPENI 1928 error code ("DSO-TYPE Not Implemented") indicates that the receiver 1929 does implement DNS Stateful Operations, but does not implement the 1930 specific DSO-TYPE of the primary TLV in the DSO request message. 1932 9.3. DSO Type Code Registry 1934 The IANA is requested to create the 16-bit DSO Type Code Registry, 1935 with initial (hexadecimal) values as shown below: 1937 +-----------+--------------------------------+----------+-----------+ 1938 | Type | Name | Status | Reference | 1939 +-----------+--------------------------------+----------+-----------+ 1940 | 0000 | Reserved | Standard | RFC-TBD | 1941 | | | | | 1942 | 0001 | KeepAlive | Standard | RFC-TBD | 1943 | | | | | 1944 | 0002 | RetryDelay | Standard | RFC-TBD | 1945 | | | | | 1946 | 0003 | EncryptionPadding | Standard | RFC-TBD | 1947 | | | | | 1948 | 0004-003F | Unassigned, reserved for | | | 1949 | | DSO session-management TLVs | | | 1950 | | | | | 1951 | 0040-F7FF | Unassigned | | | 1952 | | | | | 1953 | F800-FBFF | Reserved for | | | 1954 | | experimental/local use | | | 1955 | | | | | 1956 | FC00-FFFF | Reserved for future expansion | | | 1957 +-----------+--------------------------------+----------+-----------+ 1959 DSO Type Code zero is reserved and is not currently intended for 1960 allocation. 1962 Registrations of new DSO Type Codes in the "Reserved for DSO session- 1963 management" range 0004-003F and the "Reserved for future expansion" 1964 range FC00-FFFF require publication of an IETF Standards Action 1965 document [RFC8126]. 1967 Requests to register additional new DSO Type Codes in the 1968 "Unassigned" range 0040-F7FF are to be recorded by IANA after Expert 1969 Review [RFC8126]. At the time of publication of this document, the 1970 Designated Expert for the newly created DSO Type Code registry is 1971 [*TBD*]. 1973 DSO Type Codes in the "experimental/local" range F800-FBFF may be 1974 used as Experimental Use or Private Use values [RFC8126] and may be 1975 used freely for development purposes, or for other purposes within a 1976 single site. No attempt is made to prevent multiple sites from using 1977 the same value in different (and incompatible) ways. There is no 1978 need for IANA to review such assignments (since IANA does not record 1979 them) and assignments are not generally useful for broad 1980 interoperability. It is the responsibility of the sites making use 1981 of "experimental/local" values to ensure that no conflicts occur 1982 within the intended scope of use. 1984 10. Security Considerations 1986 If this mechanism is to be used with DNS over TLS, then these 1987 messages are subject to the same constraints as any other DNS-over- 1988 TLS messages and MUST NOT be sent in the clear before the TLS session 1989 is established. 1991 The data field of the "Encryption Padding" TLV could be used as a 1992 covert channel. 1994 When designing new DSO TLVs, the potential for data in the TLV to be 1995 used as a tracking identifier should be taken into consideration, and 1996 should be avoided when not required. 1998 When used without TLS or similar cryptographic protection, a 1999 malicious entity maybe able to inject a malicious Retry Delay 2000 Unacknowledged Message into the data stream, specifying an 2001 unreasonably large RETRY DELAY, causing a denial-of-service attack 2002 against the client. 2004 11. Acknowledgements 2006 Thanks to Stephane Bortzmeyer, Tim Chown, Ralph Droms, Paul Hoffman, 2007 Jan Komissar, Edward Lewis, Allison Mankin, Rui Paulo, David 2008 Schinazi, Manju Shankar Rao, and Bernie Volz for their helpful 2009 contributions to this document. 2011 12. References 2013 12.1. Normative References 2015 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 2016 STD 13, RFC 1034, DOI 10.17487/RFC1034, November 1987, 2017 . 2019 [RFC1035] Mockapetris, P., "Domain names - implementation and 2020 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 2021 November 1987, . 2023 [RFC1918] Rekhter, Y., Moskowitz, B., Karrenberg, D., de Groot, G., 2024 and E. Lear, "Address Allocation for Private Internets", 2025 BCP 5, RFC 1918, DOI 10.17487/RFC1918, February 1996, 2026 . 2028 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2029 Requirement Levels", BCP 14, RFC 2119, 2030 DOI 10.17487/RFC2119, March 1997, 2031 . 2033 [RFC2132] Alexander, S. and R. Droms, "DHCP Options and BOOTP Vendor 2034 Extensions", RFC 2132, DOI 10.17487/RFC2132, March 1997, 2035 . 2037 [RFC2136] Vixie, P., Ed., Thomson, S., Rekhter, Y., and J. Bound, 2038 "Dynamic Updates in the Domain Name System (DNS UPDATE)", 2039 RFC 2136, DOI 10.17487/RFC2136, April 1997, 2040 . 2042 [RFC5382] Guha, S., Ed., Biswas, K., Ford, B., Sivakumar, S., and P. 2043 Srisuresh, "NAT Behavioral Requirements for TCP", BCP 142, 2044 RFC 5382, DOI 10.17487/RFC5382, October 2008, 2045 . 2047 [RFC6891] Damas, J., Graff, M., and P. Vixie, "Extension Mechanisms 2048 for DNS (EDNS(0))", STD 75, RFC 6891, 2049 DOI 10.17487/RFC6891, April 2013, 2050 . 2052 [RFC7766] Dickinson, J., Dickinson, S., Bellis, R., Mankin, A., and 2053 D. Wessels, "DNS Transport over TCP - Implementation 2054 Requirements", RFC 7766, DOI 10.17487/RFC7766, March 2016, 2055 . 2057 [RFC7828] Wouters, P., Abley, J., Dickinson, S., and R. Bellis, "The 2058 edns-tcp-keepalive EDNS0 Option", RFC 7828, 2059 DOI 10.17487/RFC7828, April 2016, 2060 . 2062 [RFC7830] Mayrhofer, A., "The EDNS(0) Padding Option", RFC 7830, 2063 DOI 10.17487/RFC7830, May 2016, 2064 . 2066 [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for 2067 Writing an IANA Considerations Section in RFCs", BCP 26, 2068 RFC 8126, DOI 10.17487/RFC8126, June 2017, 2069 . 2071 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2072 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 2073 May 2017, . 2075 12.2. Informative References 2077 [I-D.ietf-dnssd-mdns-relay] 2078 Lemon, T. and S. Cheshire, "Multicast DNS Discovery 2079 Relay", draft-ietf-dnssd-mdns-relay-00 (work in progress), 2080 May 2018. 2082 [I-D.ietf-dnssd-push] 2083 Pusateri, T. and S. Cheshire, "DNS Push Notifications", 2084 draft-ietf-dnssd-push-14 (work in progress), March 2018. 2086 [I-D.ietf-dprive-padding-policy] 2087 Mayrhofer, A., "Padding Policy for EDNS(0)", draft-ietf- 2088 dprive-padding-policy-05 (work in progress), April 2018. 2090 [I-D.ietf-tls-tls13] 2091 Rescorla, E., "The Transport Layer Security (TLS) Protocol 2092 Version 1.3", draft-ietf-tls-tls13-28 (work in progress), 2093 March 2018. 2095 [NagleDA] Cheshire, S., "TCP Performance problems caused by 2096 interaction between Nagle's Algorithm and Delayed ACK", 2097 May 2005, 2098 . 2100 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 2101 DOI 10.17487/RFC0768, August 1980, 2102 . 2104 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 2105 Communication Layers", STD 3, RFC 1122, 2106 DOI 10.17487/RFC1122, October 1989, 2107 . 2109 [RFC2782] Gulbrandsen, A., Vixie, P., and L. Esibov, "A DNS RR for 2110 specifying the location of services (DNS SRV)", RFC 2782, 2111 DOI 10.17487/RFC2782, February 2000, 2112 . 2114 [RFC6335] Cotton, M., Eggert, L., Touch, J., Westerlund, M., and S. 2115 Cheshire, "Internet Assigned Numbers Authority (IANA) 2116 Procedures for the Management of the Service Name and 2117 Transport Protocol Port Number Registry", BCP 165, 2118 RFC 6335, DOI 10.17487/RFC6335, August 2011, 2119 . 2121 [RFC6762] Cheshire, S. and M. Krochmal, "Multicast DNS", RFC 6762, 2122 DOI 10.17487/RFC6762, February 2013, 2123 . 2125 [RFC6763] Cheshire, S. and M. Krochmal, "DNS-Based Service 2126 Discovery", RFC 6763, DOI 10.17487/RFC6763, February 2013, 2127 . 2129 [RFC7413] Cheng, Y., Chu, J., Radhakrishnan, S., and A. Jain, "TCP 2130 Fast Open", RFC 7413, DOI 10.17487/RFC7413, December 2014, 2131 . 2133 [RFC7858] Hu, Z., Zhu, L., Heidemann, J., Mankin, A., Wessels, D., 2134 and P. Hoffman, "Specification for DNS over Transport 2135 Layer Security (TLS)", RFC 7858, DOI 10.17487/RFC7858, May 2136 2016, . 2138 Authors' Addresses 2140 Ray Bellis 2141 Internet Systems Consortium, Inc. 2142 950 Charter Street 2143 Redwood City CA 94063 2144 USA 2146 Phone: +1 (650) 423-1200 2147 Email: ray@isc.org 2148 Stuart Cheshire 2149 Apple Inc. 2150 One Apple Park Way 2151 Cupertino CA 95014 2152 USA 2154 Phone: +1 (408) 996-1010 2155 Email: cheshire@apple.com 2157 John Dickinson 2158 Sinodun Internet Technologies 2159 Magadalen Centre 2160 Oxford Science Park 2161 Oxford OX4 4GA 2162 United Kingdom 2164 Email: jad@sinodun.com 2166 Sara Dickinson 2167 Sinodun Internet Technologies 2168 Magadalen Centre 2169 Oxford Science Park 2170 Oxford OX4 4GA 2171 United Kingdom 2173 Email: sara@sinodun.com 2175 Ted Lemon 2176 Barefoot Consulting 2177 Brattleboro 2178 VT 05301 2179 USA 2181 Email: mellon@fugue.com 2183 Tom Pusateri 2184 Unaffiliated 2185 Raleigh NC 27608 2186 USA 2188 Phone: +1 (919) 867-1330 2189 Email: pusateri@bangj.com