idnits 2.17.1 draft-ietf-siprec-protocol-11.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (October 17, 2013) is 3836 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(See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Outdated reference: A later version (-22) exists of draft-ietf-siprec-metadata-12 == Outdated reference: A later version (-12) exists of draft-ietf-siprec-architecture-08 Summary: 0 errors (**), 0 flaws (~~), 3 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 SIPREC L. Portman 3 Internet-Draft NICE Systems 4 Intended status: Standards Track H. Lum, Ed. 5 Expires: April 20, 2014 Genesys 6 C. Eckel 7 Cisco 8 A. Johnston 9 Avaya 10 A. Hutton 11 Unify 12 October 17, 2013 14 Session Recording Protocol 15 draft-ietf-siprec-protocol-11 17 Abstract 19 This document specifies the use of the Session Initiation Protocol 20 (SIP), the Session Description Protocol (SDP), and the Real Time 21 Protocol (RTP) for delivering real-time media and metadata from a 22 Communication Session (CS) to a recording device. The Session 23 Recording Protocol specifies the use of SIP, SDP, and RTP to 24 establish a Recording Session (RS) between the Session Recording 25 Client (SRC), which is on the path of the CS, and a Session Recording 26 Server (SRS) at the recording device. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at http://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on April 20, 2014. 45 Copyright Notice 47 Copyright (c) 2013 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (http://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 63 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 64 3. Definitions . . . . . . . . . . . . . . . . . . . . . . . . . 4 65 4. Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 66 5. Overview of operations . . . . . . . . . . . . . . . . . . . 5 67 5.1. Delivering recorded media . . . . . . . . . . . . . . . . 5 68 5.2. Delivering recording metadata . . . . . . . . . . . . . . 7 69 5.3. Receiving recording indications and providing recording 70 preferences . . . . . . . . . . . . . . . . . . . . . . . 8 71 6. SIP Handling . . . . . . . . . . . . . . . . . . . . . . . . 9 72 6.1. Procedures at the SRC . . . . . . . . . . . . . . . . . . 9 73 6.1.1. Initiating a Recording Session . . . . . . . . . . . 9 74 6.1.2. SIP extensions for recording indication and 75 preference . . . . . . . . . . . . . . . . . . . . . 10 76 6.2. Procedures at the SRS . . . . . . . . . . . . . . . . . . 10 77 6.3. Procedures for Recording-aware User Agents . . . . . . . 11 78 7. SDP Handling . . . . . . . . . . . . . . . . . . . . . . . . 11 79 7.1. Procedures at the SRC . . . . . . . . . . . . . . . . . . 12 80 7.1.1. SDP handling in RS . . . . . . . . . . . . . . . . . 12 81 7.1.1.1. Handling media stream updates . . . . . . . . . . 13 82 7.1.2. Recording indication in CS . . . . . . . . . . . . . 13 83 7.1.3. Recording preference in CS . . . . . . . . . . . . . 14 84 7.2. Procedures at the SRS . . . . . . . . . . . . . . . . . . 14 85 7.3. Procedures for Recording-aware User Agents . . . . . . . 16 86 7.3.1. Recording indication . . . . . . . . . . . . . . . . 17 87 7.3.2. Recording preference . . . . . . . . . . . . . . . . 17 88 8. RTP Handling . . . . . . . . . . . . . . . . . . . . . . . . 18 89 8.1. RTP Mechanisms . . . . . . . . . . . . . . . . . . . . . 18 90 8.1.1. RTCP . . . . . . . . . . . . . . . . . . . . . . . . 18 91 8.1.2. RTP Profile . . . . . . . . . . . . . . . . . . . . . 19 92 8.1.3. SSRC . . . . . . . . . . . . . . . . . . . . . . . . 19 93 8.1.4. CSRC . . . . . . . . . . . . . . . . . . . . . . . . 20 94 8.1.5. SDES . . . . . . . . . . . . . . . . . . . . . . . . 20 95 8.1.5.1. CNAME . . . . . . . . . . . . . . . . . . . . . . 20 96 8.1.6. Keepalive . . . . . . . . . . . . . . . . . . . . . . 20 97 8.1.7. RTCP Feedback Messages . . . . . . . . . . . . . . . 21 98 8.1.7.1. Full Intra Request . . . . . . . . . . . . . . . 21 99 8.1.7.2. Picture Loss Indicator . . . . . . . . . . . . . 21 100 8.1.7.3. Temporary Maximum Media Stream Bit Rate Request . 22 101 8.1.8. Symmetric RTP/RTCP for Sending and Receiving . . . . 22 102 8.2. Roles . . . . . . . . . . . . . . . . . . . . . . . . . . 23 103 8.2.1. SRC acting as an RTP Translator . . . . . . . . . . . 24 104 8.2.1.1. Forwarding Translator . . . . . . . . . . . . . . 24 105 8.2.1.2. Transcoding Translator . . . . . . . . . . . . . 24 106 8.2.2. SRC acting as an RTP Mixer . . . . . . . . . . . . . 25 107 8.2.3. SRC acting as an RTP Endpoint . . . . . . . . . . . . 26 108 8.3. RTP Session Usage by SRC . . . . . . . . . . . . . . . . 26 109 8.3.1. SRC Using Multiple m-lines . . . . . . . . . . . . . 26 110 8.3.2. SRC Using Mixing . . . . . . . . . . . . . . . . . . 27 111 8.4. RTP Session Usage by SRS . . . . . . . . . . . . . . . . 28 112 9. Metadata . . . . . . . . . . . . . . . . . . . . . . . . . . 29 113 9.1. Procedures at the SRC . . . . . . . . . . . . . . . . . . 29 114 9.2. Procedures at the SRS . . . . . . . . . . . . . . . . . . 31 115 9.2.1. Formal Syntax . . . . . . . . . . . . . . . . . . . . 32 116 10. Persistent Recording . . . . . . . . . . . . . . . . . . . . 32 117 11. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 33 118 11.1. Registration of Option Tags . . . . . . . . . . . . . . 33 119 11.1.1. siprec Option Tag . . . . . . . . . . . . . . . . . 33 120 11.1.2. record-aware Option Tag . . . . . . . . . . . . . . 33 121 11.2. Registration of media feature tags . . . . . . . . . . . 34 122 11.2.1. src feature tag . . . . . . . . . . . . . . . . . . 34 123 11.2.2. srs feature tag . . . . . . . . . . . . . . . . . . 34 124 11.3. New Content-Disposition Parameter Registrations . . . . 35 125 11.4. Media Type Registration . . . . . . . . . . . . . . . . 35 126 11.4.1. Registration of MIME Type application/rs-metadata . 35 127 11.4.2. Registration of MIME Type application/rs-metadata- 128 request . . . . . . . . . . . . . . . . . . . . . . 35 129 11.5. SDP Attributes . . . . . . . . . . . . . . . . . . . . . 35 130 11.5.1. 'record' SDP Attribute . . . . . . . . . . . . . . . 35 131 11.5.2. 'recordpref' SDP Attribute . . . . . . . . . . . . . 36 132 12. Security Considerations . . . . . . . . . . . . . . . . . . . 36 133 12.1. Authentication and Authorization . . . . . . . . . . . . 37 134 12.2. RTP handling . . . . . . . . . . . . . . . . . . . . . . 37 135 12.3. Metadata . . . . . . . . . . . . . . . . . . . . . . . . 38 136 12.4. Storage and playback . . . . . . . . . . . . . . . . . . 38 137 13. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 38 138 14. References . . . . . . . . . . . . . . . . . . . . . . . . . 39 139 14.1. Normative References . . . . . . . . . . . . . . . . . . 39 140 14.2. Informative References . . . . . . . . . . . . . . . . . 39 141 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 41 143 1. Introduction 144 This document specifies the mechanism to record a Communication 145 Session (CS) by delivering real-time media and metadata from the CS 146 to a recording device. In accordance to the architecture 147 [I-D.ietf-siprec-architecture], the Session Recording Protocol 148 specifies the use of SIP, SDP, and RTP to establish a Recording 149 Session (RS) between the Session Recording Client (SRC), which is on 150 the path of the CS, and a Session Recording Server (SRS) at the 151 recording device. 153 SIP is also used to deliver metadata to the recording device, as 154 specified in [I-D.ietf-siprec-metadata]. Metadata is information 155 that describes recorded media and the CS to which they relate. 157 The Session Recording Protocol intends to satisfy the SIP-based Media 158 Recording requirements listed in [RFC6341]. 160 In addition to the Session Recording Protocol, this document 161 specifies extensions for user agents that are participants in a CS to 162 receive recording indications and to provide preferences for 163 recording. 165 2. Terminology 167 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 168 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 169 document are to be interpreted as described in [RFC2119]. 171 3. Definitions 173 This document refers to the core definitions provided in the 174 architecture document [I-D.ietf-siprec-architecture]. 176 The RTP Handling section uses the definitions provided in "RTP: A 177 Transport Protocol for Real-Time Application" [RFC3550]. 179 4. Scope 181 The scope of the Session Recording Protocol includes the 182 establishment of the recording sessions and the reporting of the 183 metadata. The scope also includes extensions supported by User 184 Agents participating in the CS such as indication of recording. The 185 user agents need not be recording-aware in order to participate in a 186 CS being recorded. 188 The following items, which are not an exhaustive list, do not 189 represent the protocol itself and are considered out of the scope of 190 the Session Recording Protocol: 192 o Delivering recorded media in real-time as the CS media 194 o Specifications of criteria to select a specific CS to be recorded 195 or triggers to record a certain CS in the future 197 o Recording policies that determine whether the CS should be 198 recorded and whether parts of the CS are to be recorded 200 o Retention policies that determine how long a recording is stored 202 o Searching and accessing the recorded media and metadata 204 o Policies governing how CS users are made aware of recording 206 o Delivering additional recording session metadata through non-SIP 207 mechanism 209 5. Overview of operations 211 This section is informative and provides a description of recording 212 operations. 214 Section 6 describes the SIP communication in a recording session 215 between a SRC and a SRS, and the procedures for recording-aware user 216 agents participating in a CS. Section 7 describes the SDP in a 217 recording session, and the procedures for recording indications and 218 recording preferences. Section 8 describes the RTP handling in a 219 recording session. Section 9 describes the mechanism to deliver 220 recording metadata from the SRC to the SRS. 222 As mentioned in the architecture document 223 [I-D.ietf-siprec-architecture], there are a number of types of call 224 flows based on the location of the Session Recording Client. The 225 following sample call flows provide a quick overview of the 226 operations between the SRC and the SRS. 228 5.1. Delivering recorded media 230 When a SIP Back-to-back User Agent (B2BUA) with SRC functionality 231 routes a call from UA(A) to UA(B), the SRC has access to the media 232 path between the user agents. When the SRC is aware that it should 233 be recording the conversation, the SRC can cause the B2BUA to bridge 234 the media between UA(A) and UA(B). The SRC then establishes the 235 Recording Session with the SRS and sends replicated media towards the 236 SRS. 238 An endpoint may also have SRC functionality, where the endpoint 239 itself establishes the Recording Session to the SRS. Since the 240 endpoint has access to the media in the Communication Session, the 241 endpoint can send replicated media towards the SRS. 243 The following is a sample call flow that shows the SRC establishing a 244 recording session towards the SRS. The call flow is essentially 245 identical when the SRC is a B2BUA or as the endpoint itself. Note 246 that the SRC can choose when to establish the Recording Session 247 independent of the Communication Session, even though the following 248 call flow suggests that the SRC is establishing the Recording Session 249 (message #5) after the Communication Session is established. 251 UA A SRC UA B SRS 252 |(1)CS INVITE | | | 253 |------------->| | | 254 | |(2)CS INVITE | | 255 | |---------------------->| | 256 | | (3) 200 OK | | 257 | |<----------------------| | 258 | (4) 200 OK | | | 259 |<-------------| | | 260 | |(5)RS INVITE with SDP | | 261 | |--------------------------------------------->| 262 | | | (6) 200 OK with SDP | 263 | |<---------------------------------------------| 264 |(7)CS RTP | | | 265 |=============>|======================>| | 266 |<=============|<======================| | 267 | |(8)RS RTP | | 268 | |=============================================>| 269 | |=============================================>| 270 |(9)CS BYE | | | 271 |------------->| | | 272 | |(10)CS BYE | | 273 | |---------------------->| | 274 | |(11)RS BYE | | 275 | |--------------------------------------------->| 276 | | | | 278 Figure 1: Basic recording call flow 280 The above call flow can also apply to the case of a centralized 281 conference with a mixer. For clarity, ACKs to INVITEs and 200 OKs to 282 BYEs are not shown. The conference focus can provide the SRC 283 functionality since the conference focus has access to all the media 284 from each conference participant. When a recording is requested, the 285 SRC delivers the metadata and the media streams to the SRS. Since 286 the conference focus has access to a mixer, the SRC may choose to mix 287 the media streams from all participants as a single mixed media 288 stream towards the SRS. 290 An SRC can use a single recording session to record multiple 291 communication sessions. Every time the SRC wants to record a new 292 call, the SRC updates the recording session with a new SDP offer to 293 add new recorded streams to the recording session, and 294 correspondingly also update the metadata for the new call. 296 An SRS can also establish a recording session to an SRC, although it 297 is beyond the scope of this document to define how an SRS would 298 specify which calls to record. 300 5.2. Delivering recording metadata 302 The SRC is responsible for the delivery of metadata to the SRS. The 303 SRC may provide an initial metadata snapshot about recorded media 304 streams in the initial INVITE content in the recording session. 305 Subsequent metadata updates can be represented as a stream of events 306 in UPDATE or reINVITE requests sent by the SRC. These metadata 307 updates are normally incremental updates to the initial metadata 308 snapshot to optimize on the size of updates, however, the SRC may 309 also decide to send a new metadata snapshot anytime. 311 Metadata is transported in the body of INVITE or UPDATE messages. 312 Certain metadata, such as the attributes of the recorded media stream 313 are located in the SDP of the recording session. 315 The SRS has the ability to send a request to the SRC to request for a 316 new metadata snapshot update from the SRC. This can happen when the 317 SRS fails to understand the current stream of incremental updates for 318 whatever reason, for example, when SRS loses the current state due to 319 internal failure. The SRS may optionally attach a reason along with 320 the snapshot request. This request allows both SRC and SRS to 321 synchronize the states with a new metadata snapshot so that further 322 metadata incremental updates will be based on the latest metadata 323 snapshot. Similar to the metadata content, the metadata snapshot 324 request is transported as content in UPDATE or INVITE sent by the SRS 325 in the recording session. 327 SRC SRS 328 | | 329 |(1) INVITE (metadata snapshot) | 330 |---------------------------------------------------->| 331 | (2)200 OK | 332 |<----------------------------------------------------| 333 |(3) ACK | 334 |---------------------------------------------------->| 335 |(4) RTP | 336 |====================================================>| 337 |====================================================>| 338 |(5) UPDATE (metadata update 1) | 339 |---------------------------------------------------->| 340 | (6) 200 OK | 341 |<----------------------------------------------------| 342 |(7) UPDATE (metadata update 2) | 343 |---------------------------------------------------->| 344 | (8) 200 OK | 345 |<----------------------------------------------------| 346 | (9) UPDATE (metadata snapshot request) | 347 |<----------------------------------------------------| 348 | (10) 200 OK | 349 |---------------------------------------------------->| 350 | (11) INVITE (metadata snapshot 2 + SDP offer) | 351 |---------------------------------------------------->| 352 | (12) 200 OK (SDP answer) | 353 |<----------------------------------------------------| 354 | (13) UPDATE (metadata update 1 based on snapshot 2) | 355 |---------------------------------------------------->| 356 | (14) 200 OK | 357 |<----------------------------------------------------| 359 Figure 2: Delivering metadata via SIP UPDATE 361 5.3. Receiving recording indications and providing recording 362 preferences 364 The SRC is responsible to provide recording indications to the 365 participants in the CS. A recording-aware UA supports receiving 366 recording indications via the SDP attribute a=record, and it can 367 specify a recording preference in the CS by including the SDP 368 attribute a=recordpref. The recording attribute is a declaration by 369 the SRC in the CS to indicate whether recording is taking place. The 370 recording preference attribute is a declaration by the recording- 371 aware UA in the CS to indicate the recording preference. 373 To illustrate how the attributes are used, if a UA (A) is initiating 374 a call to UA (B) and UA (A) is also an SRC that is performing the 375 recording, then UA (A) provides the recording indication in the SDP 376 offer with a=record:on. Since UA (A) is the SRC, UA (A) receives the 377 recording indication from the SRC directly. When UA (B) receives the 378 SDP offer, UA (B) will see that recording is happening on the other 379 endpoint of this session. Since UA (B) is not an SRC and does not 380 provide any recording preference, the SDP answer does not contain 381 a=record nor a=recordpref. 383 UA A UA B 384 (SRC) | 385 | | 386 | [SRC recording starts] | 387 |(1) INVITE (SDP offer + a=record:on) | 388 |---------------------------------------------------->| 389 | (2) 200 OK (SDP answer) | 390 |<----------------------------------------------------| 391 |(3) ACK | 392 |---------------------------------------------------->| 393 |(4) RTP | 394 |<===================================================>| 395 | | 396 | [UA B wants to set preference to no recording] | 397 | (5) INVITE (SDP offer + a=recordpref:off) | 398 |<----------------------------------------------------| 399 | [SRC honors the preference and stops recording] | 400 |(6) 200 OK (SDP answer + a=record:off) | 401 |---------------------------------------------------->| 402 | (7) ACK | 403 |<----------------------------------------------------| 405 Figure 3: Recording indication and recording preference 407 After the call is established and recording is in progress, UA (B) 408 later decides to change the recording preference to no recording and 409 sends a reINVITE with the a=recordpref attribute. It is up to the 410 SRC to honor the preference, and in this case SRC decides to stop the 411 recording and updates the recording indication in the SDP answer. 413 6. SIP Handling 415 6.1. Procedures at the SRC 417 6.1.1. Initiating a Recording Session 419 A recording session is a SIP session with specific extensions 420 applied, and these extensions are listed in the procedures for SRC 421 and SRS below. When an SRC or an SRS receives a SIP session that is 422 not a recording session, it is up to the SRC or the SRS to determine 423 what to do with the SIP session. 425 The SRC can initiate a recording session by sending a SIP INVITE 426 request to the SRS. The SRC and the SRS are identified in the From 427 and To headers, respectively. 429 The SRC MUST include the '+sip.src' feature tag in the Contact URI, 430 defined in this specification as an extension to [RFC3840], for all 431 recording sessions. An SRS uses the presence of the '+sip.src' 432 feature tag in dialog creating and modifying requests and responses 433 to confirm that the dialog being created is for the purpose of a 434 Recording Session. In addition, when an SRC sends a REGISTER request 435 to a registrar, the SRC MAY include the '+sip.src' feature tag to 436 indicate the that it is a SRC. 438 Since SIP Caller Preferences extensions are optional to implement for 439 routing proxies, there is no guarantee that a recording session will 440 be routed to an SRC or SRS. A new options tag is introduced: 441 "siprec". As per [RFC3261], only an SRC or an SRS can accept this 442 option tag in a recording session. An SRC MUST include the "siprec" 443 option tag in the Require header when initiating a Recording Session 444 so that UA's which do not support the session recording protocol 445 extensions will simply reject the INVITE request with a 420 Bad 446 Extension. 448 When an SRC receives a new INVITE, the SRC MUST only consider the SIP 449 session as a recording session when both the '+sip.srs' feature tag 450 and 'siprec' option tag are included in the INVITE request. 452 6.1.2. SIP extensions for recording indication and preference 454 For the communication session, the SRC MUST provide recording 455 indication to all participants in the CS. A participant UA in a CS 456 can indicate that it is recording-aware by providing the "record- 457 aware" option tag, and the SRC MUST provide recording indications in 458 the new SDP a=record attribute described in the SDP Handling section. 459 In the absence of the "record-aware" option tag, meaning that the 460 participant UA is not recording-aware, an SRC MUST provide recording 461 indications through other means such as playing a tone inband, if the 462 SRC is required to do so (e.g. based on policies). 464 An SRC in the CS may also indicate itself as a session recording 465 client by including the '+sip.src' feature tag. A recording-aware 466 participant can learn that a SRC is in the CS, and can set the 467 recording preference for the CS with the new SDP a=recordpref 468 attribute described in the SDP Handling section below. 470 6.2. Procedures at the SRS 471 When an SRS receives a new INVITE, the SRS MUST only consider the SIP 472 session as a recording session when both the '+sip.src' feature tag 473 and 'siprec' option tag are included in the INVITE request. 475 The SRS can initiate a recording session by sending a SIP INVITE 476 request to the SRC. The SRS and the SRC are identified in the From 477 and To headers, respectively. 479 The SRS MUST include the '+sip.srs' feature tag in the Contact URI, 480 as per [RFC3840], for all recording sessions. An SRC uses the 481 presence of this feature tag in dialog creating and modifying 482 requests and responses to confirm that the dialog being created is 483 for the purpose of a Recording Session (REQ-30). In addition, when 484 an SRS sends a REGISTER request to a registrar, the SRS SHOULD 485 include the '+sip.srs' feature tag to indicate that it is a SRS. 487 An SRS MUST include the "siprec" option tag in the Require header as 488 per [RFC3261] when initiating a Recording Session so that UA's which 489 do not support the session recording protocol extensions will simply 490 reject the INVITE request with a 420 Bad Extension. 492 6.3. Procedures for Recording-aware User Agents 494 A recording-aware user agent is a participant in the CS that supports 495 the SIP and SDP extensions for receiving recording indication and for 496 requesting recording preferences for the call. A recording-aware UA 497 MUST indicate that it can accept reporting of recording indication 498 provided by the SRC with a new option tag "record-aware" when 499 initiating or establishing a CS, meaning including the "record-aware" 500 tag in the Supported header in the initial INVITE request or 501 response. 503 A recording-aware UA MUST be prepared to provide a recording 504 indication to the end user through an appropriate user interface, 505 indicating whether recording is on, off, or paused for each medium. 506 Some user agents that are automatons (e.g. IVR, media server, PSTN 507 gateway) may not have a user interface to render recording 508 indication. When such user agent indicates recording awareness, the 509 UA SHOULD render recording indication through other means, such as 510 passing an inband tone on the PSTN gateway, putting the recording 511 indication in a log file, or raising an application event in a 512 VoiceXML dialog. These user agents MAY also choose not to indicate 513 recording awareness, thereby relying on whatever mechanism an SRC 514 chooses to indicate recording, such as playing a tone inband. 516 7. SDP Handling 517 7.1. Procedures at the SRC 519 The SRC and SRS follows the SDP offer/answer model in [RFC3264]. The 520 procedures for SRC and SRS describe the conventions used in a 521 recording session. 523 7.1.1. SDP handling in RS 525 Since the SRC does not expect to receive media from the SRS, the SRC 526 typically sets each media stream of the SDP offer to only send media, 527 by qualifying them with the a=sendonly attribute, according to the 528 procedures in [RFC3264]. 530 The SRC sends recorded streams of participants to the SRS, and the 531 SRC MUST provide a label attribute (a=label), as per [RFC4574], on 532 each media stream in order to identify the recorded stream with the 533 rest of the metadata. The a=label attribute identifies each recorded 534 media stream, and the label name is mapped to the Media Stream 535 Reference in the metadata as per [I-D.ietf-siprec-metadata]. The 536 scope of the a=label attribute only applies to the SDP and Metadata 537 conveyed in the bodies of the SIP request or response that the label 538 appeared in. Note that a recorded stream is distinct from a CS 539 stream; the metadata provides a list of participants that contributes 540 to each recorded stream. 542 The following is an example SDP offer from SRC with both audio and 543 video recorded streams. Note that the following example contains 544 unfolded lines longer than 72 characters. These are captured between 545 tags. 547 v=0 548 o=SRC 2890844526 2890844526 IN IP4 198.51.100.1 549 s=- 550 c=IN IP4 198.51.100.1 551 t=0 0 552 m=audio 12240 RTP/AVP 0 4 8 553 a=sendonly 554 a=label:1 555 m=video 22456 RTP/AVP 98 556 a=rtpmap:98 H264/90000 557 558 a=fmtp:98 profile-level-id=42A01E; 559 sprop-parameter-sets=Z0IACpZTBYmI,aMljiA== 560 561 a=sendonly 562 a=label:2 563 m=audio 12242 RTP/AVP 0 4 8 564 a=sendonly 565 a=label:3 566 m=video 22458 RTP/AVP 98 567 a=rtpmap:98 H264/90000 568 569 a=fmtp:98 profile-level-id=42A01E; 570 sprop-parameter-sets=Z0IACpZTBYmI,aMljiA== 571 572 a=sendonly 573 a=label:4 575 Figure 4: Sample SDP offer from SRC with audio and video streams 577 7.1.1.1. Handling media stream updates 579 Over the lifetime of a recording session, the SRC can add and remove 580 recorded streams from the recording session for various reasons. For 581 example, when a CS stream is added or removed from the CS, or when a 582 CS is created or terminated if a recording session handles multiple 583 CSes. To remove a recorded stream from the recording session, the 584 SRC sends a new SDP offer where the port of the media stream to be 585 removed is set to zero, according to the procedures in [RFC3264]. To 586 add a recorded stream to the recording session, the SRC sends a new 587 SDP offer by adding a new media stream description or by reusing an 588 old media stream which had been previously disabled, according to the 589 procedures in [RFC3264]. 591 The SRC can temporarily discontinue streaming and collection of 592 recorded media from the SRC to the SRS for reasons such as masking 593 the recording. In this case, the SRC sends a new SDP offer and sets 594 the media stream to inactive (a=inactive) for each recorded stream to 595 be paused, as per the procedures in [RFC3264]. To resume streaming 596 and collection of recorded media, the SRC sends a new SDP offer and 597 sets the media stream to sendonly (a=sendonly). Note that a CS 598 itself may change the media stream direction by updating the SDP, for 599 example, by setting a=inactive for SDP hold. Media stream direction 600 changes in CS are conveyed in the metadata by the SRC. The SRC MUST 601 NOT modify the RS media stream with a=inactive for SDP hold on the CS 602 since this operation is reserved for pausing the RS media, however, 603 an SRC can have a local policy to pause the RS media when the CS is 604 placed on hold. 606 7.1.2. Recording indication in CS 608 While there are existing mechanisms for providing an indication that 609 a CS is being recorded, these mechanisms are usually delivered on the 610 CS media streams such as playing an in-band tone or an announcement 611 to the participants. A new 'record' SDP attribute is introduced to 612 allow the SRC to indicate recording state to a recording-aware UA in 613 CS. 615 The 'record' SDP attribute appears at the media level or session 616 level in either SDP offer or answer. When the attribute is applied 617 at the session level, the indication applies to all media streams in 618 the SDP. When the attribute is applied at the media level, the 619 indication applies to the media stream only, and that overrides the 620 indication if also set at the session level. Whenever the recording 621 indication needs to change, such as termination of recording, then 622 the SRC MUST initiate a reINVITE or UPDATE to update the SDP a=record 623 attribute. 625 The following is the ABNF of the 'record' attribute: 627 attribute /= record-attr 628 ; attribute defined in RFC 4566 630 record-attr = "record:" indication 631 indication = "on" / "off" / "paused" 633 on Recording is in progress. 635 off No recording is in progress. 637 paused Recording is in progress but media is paused. 639 7.1.3. Recording preference in CS 641 When the SRC receives the a=recordpref SDP in an SDP offer or answer, 642 the SRC chooses to honor the preference to record based on local 643 policy at the SRC. If the SRC makes a change in recording state, 644 then the SRC reports the recording state in the a=record attribute in 645 the SDP answer or in a subsequent SDP offer/answer. 647 7.2. Procedures at the SRS 648 Typically the SRS only receives RTP streams from the SRC; therefore, 649 the SDP offer/answer from the SRS normally sets each media stream to 650 receive media, by setting them with the a=recvonly attribute, 651 according to the procedures of [RFC3264]. When the SRS is not ready 652 to receive a recorded stream, the SRS sets the media stream as 653 inactive in the SDP offer or answer by setting it with a=inactive 654 attribute, according to the procedures of [RFC3264]. When the SRS is 655 ready to receive recorded streams, the SRS sends a new SDP offer and 656 sets the media streams with a=recvonly attribute. 658 The following is an example of SDP answer from SRS for the SDP offer 659 from the above sample. Note that the following example contain 660 unfolded lines longer than 72 characters. These are captured between 661 tags. 663 v=0 664 o=SRS 0 0 IN IP4 198.51.100.20 665 s=- 666 c=IN IP4 198.51.100.20 667 t=0 0 668 m=audio 10000 RTP/AVP 0 669 a=recvonly 670 a=label:1 671 m=video 10002 RTP/AVP 98 672 a=rtpmap:98 H264/90000 673 674 a=fmtp:98 profile-level-id=42A01E; 675 sprop-parameter-sets=Z0IACpZTBYmI,aMljiA== 676 677 a=recvonly 678 a=label:2 679 m=audio 10004 RTP/AVP 0 680 a=recvonly 681 a=label:3 682 m=video 10006 RTP/AVP 98 683 a=rtpmap:98 H264/90000 684 685 a=fmtp:98 profile-level-id=42A01E; 686 sprop-parameter-sets=Z0IACpZTBYmI,aMljiA== 687 688 a=recvonly 689 a=label:4 691 Figure 5: Sample SDP answer from SRS with audio and video streams 693 Over the lifetime of a recording session, the SRS can remove recorded 694 streams from the recording session for various reasons. To remove a 695 recorded stream from the recording session, the SRS sends a new SDP 696 offer where the port of the media stream to be removed is set to 697 zero, according to the procedures in [RFC3264]. 699 The SRS SHOULD NOT add recorded streams in the recording session when 700 SRS sends a new SDP offer. Similarly, when the SRS starts a 701 recording session, the SRS SHOULD initiate the INVITE without an SDP 702 offer to let the SRC generate the SDP offer with recorded streams. 704 The following sequence diagram shows an example where the SRS is 705 initially not ready to receive recorded streams, and later updates 706 the recording session when the SRS is ready to record. 708 SRC SRS 709 | | 710 |(1) INVITE (SDP offer) | 711 |---------------------------------------------------->| 712 | [not ready to record] 713 | (2)200 OK with SDP inactive | 714 |<----------------------------------------------------| 715 |(3) ACK | 716 |---------------------------------------------------->| 717 | ... | 718 | [ready to record] 719 | (4) re-INVITE with SDP recvonly | 720 |<----------------------------------------------------| 721 |(5)200 OK with SDP sendonly | 722 |---------------------------------------------------->| 723 | (6) ACK | 724 |<----------------------------------------------------| 725 |(7) RTP | 726 |====================================================>| 727 | ... | 728 |(8) BYE | 729 |---------------------------------------------------->| 730 | (9) OK | 731 |<----------------------------------------------------| 733 Figure 6: SRS responding to offer with a=inactive 735 7.3. Procedures for Recording-aware User Agents 736 7.3.1. Recording indication 738 When a recording-aware UA receives an SDP offer or answer that 739 includes the a=record attribute, the UA MUST provide the recording 740 indication to the end user whether the recording is on, off, or 741 paused for each medium based on the most recently received a=record 742 SDP attribute for that medium. 744 When a CS is traversed through multiple UAs such as a B2BUA or a 745 conference focus, each UA involved in the CS that is aware that the 746 CS is being recorded MUST provide the recording indication through 747 the a=record attribute to all other parties in the CS. 749 It is possible that more than one SRC can be in the call path 750 recording the same CS, but the recording indication attribute does 751 not provide any hint as to which SRC is actually performing the 752 recording. This means that an endpoint only knows that the call is 753 being recorded through the recording indicator. This attribute is 754 also not used as an indication to negotiate which SRC in the call 755 path will perform recording and is not used as a request to start/ 756 stop recording if there are multiple SRCs in the call path. 758 7.3.2. Recording preference 760 A participant in a CS MAY set the recording preference in the CS to 761 be recorded or not recorded at session establishment or during the 762 session. A new 'recordpref' SDP attribute is introduced, and the 763 participant in CS may set this recording preference attribute in any 764 SDP offer/answer at session establishment time or during the session. 765 The SRC is not required to honor the recording preference from a 766 participant based on local policies at the SRC, and the participant 767 can learn the recording indication through the a=record SDP attribute 768 as described in the above section. 770 The SDP a=recordpref attribute can appear at the media level or 771 session level and can appear in an SDP offer or answer. When the 772 attribute is applied at the session level, the recording preference 773 applies to all media stream in the SDP. When the attribute is 774 applied at the media level, the recording preference applies to the 775 media stream only, and that overrides the recording preference if 776 also set at the session level. The user agent can change the 777 recording preference by changing the a=recordpref attribute in 778 subsequent SDP offer or answer. The absence of the a=recordpref 779 attribute in the SDP indicates that the UA has no recording 780 preference. 782 The following is the ABNF of the recordpref attribute: 784 attribute /= recordpref-attr 785 ; attribute defined in RFC 4566 787 recordpref-attr = "a=recordpref:" pref 788 pref = "on" / "off" / "pause" / "nopreference" 790 on Sets the preference to record if it has not already been started. 791 If the recording is currently paused, the preference is to resume 792 recording. 794 off Sets the preference for no recording. If recording has already 795 been started, then the preference is to stop the recording. 797 pause If the recording is currently in progress, sets the preference 798 to pause the recording. 800 nopreference To indicate that the UA has no preference on recording. 802 8. RTP Handling 804 This section provides recommendations and guidelines for RTP and RTCP 805 in the context of SIPREC. In order to communicate most effectively, 806 the Session Recording Client (SRC), the Session Recording Server 807 (SRS), and any Recording aware User Agents (UAs) SHOULD utilize the 808 mechanisms provided by RTP in a well-defined and predicable manner. 809 It is the goal of this document to make the reader aware of these 810 mechanisms and provide recommendations and guidelines. 812 8.1. RTP Mechanisms 814 This section briefly describes important RTP/RTCP constructs and 815 mechanisms that are particularly useful within the content of SIPREC. 817 8.1.1. RTCP 819 The RTP data transport is augmented by a control protocol (RTCP) to 820 allow monitoring of the data delivery. RTCP, as defined in 821 [RFC3550], is based on the periodic transmission of control packets 822 to all participants in the RTP session, using the same distribution 823 mechanism as the data packets. Support for RTCP is REQUIRED, per 824 [RFC3550], and it provides, among other things, the following 825 important functionality in relation to SIPREC: 827 1) Feedback on the quality of the data distribution 829 This feedback from the receivers may be used to diagnose faults in 830 the distribution. As such, RTCP is a well-defined and efficient 831 mechanism for the SRS to inform the SRC, and for the SRC to inform 832 Recording aware UAs, of issues that arise with respect to the 833 reception of media that is to be recorded. 835 2) Carries a persistent transport-level identifier for an RTP source 836 called the canonical name or CNAME 838 The SSRC identifier may change if a conflict is discovered or a 839 program is restarted; in which case receivers can use the CNAME to 840 keep track of each participant. Receivers may also use the CNAME to 841 associate multiple data streams from a given participant in a set of 842 related RTP sessions, for example to synchronize audio and video. 843 Synchronization of media streams is also facilitated by the NTP and 844 RTP timestamps included in RTCP packets by data senders. 846 8.1.2. RTP Profile 848 The RECOMMENDED RTP profiles for the SRC, SRS, and Recording aware 849 UAs are "Extended Secure RTP Profile for Real-time Transport Control 850 Protocol (RTCP)-Based Feedback (RTP/SAVPF)", [RFC5124] when using 851 encrypted RTP streams, and "Extended RTP Profile for Real-time 852 Transport Control Protocol (RTCP)-Based Feedback (RTP/AVPF)", 853 [RFC4585] when using non encrypted media streams. However, as this 854 is not a requirement, some implementations may use "The Secure Real- 855 time Transport Protocol (SRTP)", [RFC3711] and "RTP Profile for Audio 856 and Video Conferences with Minimal Control", AVP [RFC3551]. 857 Therefore, it is RECOMMENDED that the SRC, SRS, and Recording aware 858 UAs not rely entirely on SAVPF or AVPF for core functionality that 859 may be at least partially achievable using SAVP and AVP. 861 AVPF and SAVPF provide an improved RTCP timer model that allows more 862 flexible transmission of RTCP packets in response to events, rather 863 than strictly according to bandwidth. AVPF based codec control 864 messages provide efficient mechanisms for an SRC, SRS, and Recording 865 aware UAs to handle events such as scene changes, error recovery, and 866 dynamic bandwidth adjustments. These messages are discussed in more 867 detail later in this document. 869 SAVP and SAVPF provide media encryption, integrity protection, replay 870 protection, and a limited form of source authentication. They do not 871 contain or require a specific keying mechanism. 873 8.1.3. SSRC 875 The synchronization source (SSRC), as defined in [RFC3550] is carried 876 in the RTP header and in various fields of RTCP packets. It is a 877 random 32-bit number that is required to be globally unique within an 878 RTP session. It is crucial that the number be chosen with care in 879 order that participants on the same network or starting at the same 880 time are not likely to choose the same number. Guidelines regarding 881 SSRC value selection and conflict resolution are provided in 882 [RFC3550]. 884 The SSRC may also be used to separate different sources of media 885 within a single RTP session. For this reason as well as for conflict 886 resolution, it is important that the SRC, SRS, and Recording aware 887 UAs handle changes in SSRC values and properly identify the reason of 888 the change. The CNAME values carried in RTCP facilitate this 889 identification. 891 8.1.4. CSRC 893 The contributing source (CSRC), as defined in [RFC3550], identifies 894 the source of a stream of RTP packets that has contributed to the 895 combined stream produced by an RTP mixer. The mixer inserts a list 896 of the SSRC identifiers of the sources that contributed to the 897 generation of a particular packet into the RTP header of that packet. 898 This list is called the CSRC list. It is RECOMMENDED that a SRC or 899 Recording aware UA, when acting a mixer, sets the CSRC list 900 accordingly, and that the SRC and SRS interpret the CSRC list 901 appropriately when received. 903 8.1.5. SDES 905 The Source Description (SDES), as defined in [RFC3550], contains an 906 SSRC/CSRC identifier followed by a list of zero or more items, which 907 carry information about the SSRC/CSRC. End systems send one SDES 908 packet containing their own source identifier (the same as the SSRC 909 in the fixed RTP header). A mixer sends one SDES packet containing a 910 chunk for each contributing source from which it is receiving SDES 911 information, or multiple complete SDES packets if there are more than 912 31 such sources. 914 8.1.5.1. CNAME 916 The Canonical End-Point Identifier (CNAME), as defined in [RFC3550], 917 provides the binding from the SSRC identifier to an identifier for 918 the source (sender or receiver) that remains constant. It is 919 important the SRC and Recording aware UAs generate CNAMEs 920 appropriately and that the SRC and SRS interpret and use them for 921 this purpose. Guidelines for generating CNAME values are provided in 922 "Guidelines for Choosing RTP Control Protocol (RTCP) Canonical Names 923 (CNAMEs)" [RFC7022]. 925 8.1.6. Keepalive 926 It is anticipated that media streams in SIPREC may exist in an 927 inactive state for extended periods of times for any of a number of 928 valid reasons. In order for the bindings and any pinholes in NATs/ 929 firewalls to remain active during such intervals, it is RECOMMENDED 930 that the SRC, SRS, and Recording aware UAs follow the keep-alive 931 procedure recommended in "Application Mechanism for Keeping Alive the 932 NAT Mappings Associated to RTP/RTP Control Protocol (RTCP) Flows" 933 [RFC6263] for all RTP media streams. 935 8.1.7. RTCP Feedback Messages 937 "Codec Control Messages in the RTP Audio-Visual Profile with Feedback 938 (AVPF)" [RFC5104] specifies extensions to the messages defined in 939 AVPF [RFC4585]. Support for and proper usage of these messages is 940 important to SRC, SRS, and Recording aware UA implementations. Note 941 that these messages are applicable only when using the AVFP or SAVPF 942 RTP profiles 944 8.1.7.1. Full Intra Request 946 A Full Intra Request (FIR) Command, when received by the designated 947 media sender, requires that the media sender sends a Decoder Refresh 948 Point at the earliest opportunity. Using a decoder refresh point 949 implies refraining from using any picture sent prior to that point as 950 a reference for the encoding process of any subsequent picture sent 951 in the stream. 953 Decoder refresh points, especially Intra or IDR pictures for H.264 954 video codecs, are in general several times larger in size than 955 predicted pictures. Thus, in scenarios in which the available bit 956 rate is small, the use of a decoder refresh point implies a delay 957 that is significantly longer than the typical picture duration. 959 8.1.7.1.1. SIP INFO for FIR 961 "XML Schema for Media Control" [RFC5168] defines an Extensible Markup 962 Language (XML) Schema for video fast update. Implementations are 963 discouraged from using the method described except for backward 964 compatibility purposes. Implementations SHOULD use FIR messages 965 instead. 967 8.1.7.2. Picture Loss Indicator 969 Picture Loss Indication (PLI), as defined in [RFC4585], informs the 970 encoder of the loss of an undefined amount of coded video data 971 belonging to one or more pictures. Using the FIR command to recover 972 from errors is explicitly disallowed, and instead the PLI message 973 SHOULD be used. FIR SHOULD be used only in situations where not 974 sending a decoder refresh point would render the video unusable for 975 the users. Examples where sending FIR is appropriate include a 976 multipoint conference when a new user joins the conference and no 977 regular decoder refresh point interval is established, and a video 978 switching MCU that changes streams. 980 8.1.7.3. Temporary Maximum Media Stream Bit Rate Request 982 A receiver, translator, or mixer uses the Temporary Maximum Media 983 Stream Bit Rate Request (TMMBR) to request a sender to limit the 984 maximum bit rate for a media stream to the provided value. 985 Appropriate use of TMMBR facilitates rapid adaptation to changes in 986 available bandwidth. 988 8.1.7.3.1. Renegotiation of SDP bandwidth attribute 990 If it is likely that the new value indicated by TMMBR will be valid 991 for the remainder of the session, the TMMBR sender is expected to 992 perform a renegotiation of the session upper limit using the session 993 signaling protocol. Therefore for SIPREC, implementations are 994 RECOMMENDED to use TMMBR for temporary changes, and renegotiation of 995 bandwidth via SDP offer/answer for more permanent changes. 997 8.1.8. Symmetric RTP/RTCP for Sending and Receiving 999 Within an SDP offer/answer exchange, RTP entities choose the RTP and 1000 RTCP transport addresses (i.e., IP addresses and port numbers) on 1001 which to receive packets. When sending packets, the RTP entities may 1002 use the same source port or a different source port as those signaled 1003 for receiving packets. When the transport address used to send and 1004 receive RTP is the same, it is termed "symmetric RTP" [RFC4961]. 1005 Likewise, when the transport address used to send and receive RTCP is 1006 the same, it is termed "symmetric RTCP" [RFC4961]. 1008 When sending RTP, it is REQUIRED to use symmetric RTP. When sending 1009 RTCP, it is REQUIRED to use symmetric RTCP. Although an SRS will not 1010 normally send RTP, it will send RTCP as well as receive RTP and RTCP. 1011 Likewise, although an SRC will not normally receive RTP from the SRS, 1012 it will receive RTCP as well as send RTP and RTCP. 1014 Note: Symmetric RTP and symmetric RTCP are different from RTP/RTCP 1015 multiplexing [RFC5761]. 1017 8.2. Roles 1019 An SRC has the task of gathering media from the various UAs in one or 1020 more Communication Sessions (CSs) and forwarding the information to 1021 the SRS within the context of a corresponding Recording Session (RS). 1022 There are numerous ways in which an SRC may do this, including but 1023 not limited to, appearing as a UA within a CS, or as a B2BUA between 1024 UAs within a CS. 1026 (Recording Session) +---------+ 1027 +------------SIP------->| | 1028 | +------RTP/RTCP----->| SRS | 1029 | | +-- Metadata -->| | 1030 | | | +---------+ 1031 v v | 1032 +---------+ 1033 | SRC | 1034 |---------| (Communication Session) +---------+ 1035 | |<----------SIP---------->| | 1036 | UA-A | | UA-B | 1037 | |<-------RTP/RTCP-------->| | 1038 +---------+ +---------+ 1040 Figure 7: UA as SRC 1042 (Recording Session) +---------+ 1043 +------------SIP------->| | 1044 | +------RTP/RTCP----->| SRS | 1045 | | +-- Metadata -->| | 1046 | | | +---------+ 1047 v v | 1048 +---------+ 1049 | SRC | 1050 +---------+ |---------| +---------+ 1051 | |<----SIP----->| |<----SIP----->| | 1052 | UA-A | | B2BUA | | UA-B | 1053 | |<--RTP/RTCP-->| |<--RTP/RTCP-->| | 1054 +---------+ +---------+ +---------+ 1055 |_______________________________________________| 1056 (Communication Session) 1058 Figure 8: B2BUA as SRC 1060 The following subsections define a set of roles an SRC may choose to 1061 play based on its position with respect to a UA within a CS, and an 1062 SRS within an RS. A CS and a corresponding RS are independent 1063 sessions; therefore, an SRC may play a different role within a CS 1064 than it does within the corresponding RS. 1066 8.2.1. SRC acting as an RTP Translator 1068 The SRC may act as a translator, as defined in [RFC3550]. A defining 1069 characteristic of a translator is that it forwards RTP packets with 1070 their SSRC identifier intact. There are two types of translators, 1071 one that simply forwards, and another that performs transcoding 1072 (e.g., from one codec to another) in addition to forwarding. 1074 8.2.1.1. Forwarding Translator 1076 When acting as a forwarding translator, RTP received as separate 1077 streams from different sources (e.g., from different UAs with 1078 different SSRCs) cannot be mixed by the SRC and MUST be sent 1079 separately to the SRS. All RTCP reports MUST be passed by the SRC 1080 between the UAs and the SRS, such that the UAs and SRS are able to 1081 detect any SSRC collisions. 1083 RTCP Sender Reports generated by a UA sending a stream MUST be 1084 forwarded to the SRS. RTCP Receiver Reports generated by the SRS 1085 MUST be forwarded to the relevant UA. 1087 UAs may receive multiple sets of RTCP Receiver Reports, one or more 1088 from other UAs participating in the CS, and one from the SRS 1089 participating in the RS. A Recording aware UA SHOULD be prepared to 1090 process the RTCP Receiver Reports from the SRS, whereas a recording 1091 unaware UA may discard such RTCP packets as not of relevance. 1093 If SRTP is used on both the CS and the RS, decryption and/or re- 1094 encryption may occur. For example, if different keys are used, it 1095 will occur. If the same keys are used, it need not occur. 1096 Section 12 provides additional information on SRTP and keying 1097 mechanisms. 1099 If packet loss occurs, either from the UA to the SRC or from the SRC 1100 to the SRS, the SRS SHOULD detect and attempt to recover from the 1101 loss. The SRC does not play a role in this other than forwarding the 1102 associated RTP and RTCP packets. 1104 8.2.1.2. Transcoding Translator 1105 When acting as a transcoding translator, an SRC MAY perform 1106 transcoding (e.g., from one codec to another), and this may result in 1107 a different rate of packets between what the SRC receives and what 1108 the SRC sends. As when acting as a forwarding translator, RTP 1109 received as separate streams from different sources (e.g., from 1110 different UAs with different SSRCs) cannot be mixed by the SRC and 1111 MUST be sent separately to the SRS. All RTCP reports MUST be passed 1112 by the SRC between the UAs and the SRS, such that the UAs and SRS are 1113 able to detect any SSRC collisions. 1115 RTCP Sender Reports generated by a UA sending a stream MUST be 1116 forwarded to the SRS. RTCP Receiver Reports generated by the SRS 1117 MUST be forwarded to the relevant UA. The SRC may need to manipulate 1118 the RTCP Receiver Reports to take account of any transcoding that has 1119 taken place. 1121 UAs may receive multiple sets of RTCP Receiver Reports, one or more 1122 from other UAs participating in the CS, and one from the SRS 1123 participating in the RS. A Recording aware UA SHOULD be prepared to 1124 process the RTCP Receiver Reports from the SRS, whereas a recording 1125 unaware UA may discard such RTCP packets as not of relevance. 1127 If SRTP is used on both the CS and the RS, decryption and/or re- 1128 encryption may occur. For example, if different keys are used, it 1129 will occur. If the same keys are used, it need not occur. 1130 Section 12 provides additional information on SRTP and keying 1131 mechanisms. 1133 If packet loss occurs, either from the UA to the SRC or from the SRC 1134 to the SRS, the SRS SHOULD detect and attempt to recover from the 1135 loss. The SRC does not play a role in this other than forwarding the 1136 associated RTP and RTCP packets. 1138 8.2.2. SRC acting as an RTP Mixer 1140 In the case of the SRC acting as a RTP mixer, as defined in 1141 [RFC3550], the SRC combines RTP streams from different UA and sends 1142 them towards the SRS using its own SSRC. The SSRCs from the 1143 contributing UA SHOULD be conveyed as CSRCs identifiers within this 1144 stream. The SRC may make timing adjustments among the received 1145 streams and generate its own timing on the stream sent to the SRS. 1146 Optionally an SRC acting as a mixer can perform transcoding, and can 1147 even cope with different codings received from different UAs. RTCP 1148 Sender Reports and Receiver Reports are not forwarded by an SRC 1149 acting as mixer, but there are requirements for forwarding RTCP 1150 Source Description (SDES) packets. The SRC generates its own RTCP 1151 Sender and Receiver reports toward the associated UAs and SRS. 1153 The use of SRTP between the SRC and the SRS for the RS is independent 1154 of the use of SRTP between the UAs and SRC for the CS. Section 12 1155 provides additional information on SRTP and keying mechanisms. 1157 If packet loss occurs from the UA to the SRC, the SRC SHOULD detect 1158 and attempt to recover from the loss. If packet loss occurs from the 1159 SRC to the SRS, the SRS SHOULD detect and attempt to recover from the 1160 loss. 1162 8.2.3. SRC acting as an RTP Endpoint 1164 The case of the SRC acting as an RTP endpoint, as defined in 1165 [RFC3550], is similar to the mixer case, except that the RTP session 1166 between the SRC and the SRS is considered completely independent from 1167 the RTP session that is part of the CS. The SRC can, but need not, 1168 mix RTP streams from different participants prior to sending to the 1169 SRS. RTCP between the SRC and the SRS is completely independent of 1170 RTCP on the CS. 1172 The use of SRTP between the SRC and the SRS for the RS is independent 1173 of the use of SRTP between the UAs and SRC for the CS. Section 12 1174 provides additional information on SRTP and keying mechanisms. 1176 If packet loss occurs from the UA to the SRC, the SRC SHOULD detect 1177 and attempt to recover from the loss. If packet loss occurs from the 1178 SRC to the SRS, the SRS SHOULD detect and attempt to recover from the 1179 loss. 1181 8.3. RTP Session Usage by SRC 1183 There are multiple ways that an SRC may choose to deliver recorded 1184 media to an SRS. In some cases, it may use a single RTP session for 1185 all media within the RS, whereas in others it may use multiple RTP 1186 sessions. The following subsections provide examples of basic RTP 1187 session usage by the SRC, including a discussion of how the RTP 1188 constructs and mechanisms covered previously are used. An SRC may 1189 choose to use one or more of the RTP session usages within a single 1190 RS. For the purpose of base interoperability between SRC and SRS, an 1191 SRC MUST support separate m-lines in SDP, one per CS media direction. 1192 The set of RTP session usages described is not meant to be 1193 exhaustive. 1195 8.3.1. SRC Using Multiple m-lines 1197 When using multiple m-lines, an SRC includes each m-line in an SDP 1198 offer to the SRS. The SDP answer from the SRS MUST include all 1199 m-lines, with any rejected m-lines indicated with a zero port, per 1200 [RFC3264]. Having received the answer, the SRC starts sending media 1201 to the SRS as indicated in the answer. Alternatively, if the SRC 1202 deems the level of support indicated in the answer to be 1203 unacceptable, it may initiate another SDP offer/answer exchange in 1204 which an alternative RTP session usage is negotiated. 1206 In order to preserve the mapping of media to participant within the 1207 CSs in the RS, the SRC SHOULD map each unique CNAME within the CSs to 1208 a unique CNAME within the RS. Additionally, the SRC SHOULD map each 1209 unique combination of CNAME/SSRC within the CSs to a unique CNAME/ 1210 SSRC within the RS. In doing to, the SRC may act as an RTP 1211 translator or as an RTP endpoint. 1213 The following figure illustrates a case in which each UA represents a 1214 participant contributing two RTP sessions (e.g. one for audio and one 1215 for video), each with a single SSRC. The SRC acts as an RTP 1216 translator and delivers the media to the SRS using four RTP sessions, 1217 each with a single SSRC. The CNAME and SSRC values used by the UAs 1218 within their media streams are preserved in the media streams from 1219 the SRC to the SRS. 1221 +---------+ 1222 +------------SSRC Aa--->| | 1223 | + --------SSRC Av--->| | 1224 | | +------SSRC Ba--->| SRS | 1225 | | | +---SSRC Bv--->| | 1226 | | | | +---------+ 1227 | | | | 1228 | | | | 1229 +---------+ +----------+ +---------+ 1230 | |---SSRC Aa-->| SRC |<--SSRC Ba---| | 1231 | UA-A | |(CNAME-A, | | UA-B | 1232 |(CNAME-A)|---SSRC Av-->| CNAME-B) |<--SSRC Bv---|(CNAME-B)| 1233 +---------+ +----------+ +---------+ 1235 Figure 9: SRC Using Multiple m-lines 1237 8.3.2. SRC Using Mixing 1239 When using mixing, the SRC combines RTP streams from different 1240 participants and sends them towards the SRS using its own SSRC. The 1241 SSRCs from the contributing participants SHOULD be conveyed as CSRCs 1242 identifiers. The SRC includes one m-line for each RTP session in an 1243 SDP offer to the SRS. The SDP answer from the SRS MUST include all 1244 m-lines, with any rejected m-lines indicated with the zero port, per 1245 [RFC3264]. Having received the answer, the SRC starts sending media 1246 to the SRS as indicated in the answer. 1248 In order to preserve the mapping of media to participant within the 1249 CSs in the RS, the SRC SHOULD map each unique CNAME within the CSs to 1250 a unique CNAME within the RS. Additionally, the SRC SHOULD map each 1251 unique combination of CNAME/SSRC within the CSs to a unique CNAME/ 1252 SSRC within the RS. The SRC MUST avoid SSRC collisions, rewriting 1253 SSRCs if necessary when used as CSRCs in the RS. In doing to, the 1254 SRC acts as an RTP mixer. 1256 In the event the SRS does not support this usage of CSRC values, it 1257 relies entirely on the SIPREC metadata to determine the participants 1258 included within each mixed stream. 1260 The following figure illustrates a case in which each UA represents a 1261 participant contributing two RTP sessions (e.g. one for audio and one 1262 for video), each with a single SSRC. The SRC acts as an RTP mixer 1263 and delivers the media to the SRS using two RTP sessions, mixing 1264 media from each participant into a single RTP session containing a 1265 single SSRC and two CSRCs. 1267 SSRC Sa +---------+ 1268 +-------CSRC Aa,Ba--->| | 1269 | | | 1270 | SSRC Sa | SRS | 1271 | +---CSRC Av,Bv--->| | 1272 | | +---------+ 1273 | | 1274 +----------+ 1275 +---------+ | SRC | +---------+ 1276 | |---SSRC Aa-->|(CNAME-S, |<--SSRC Ba---| | 1277 | UA-A | | CNAME-A, | | UA-B | 1278 |(CNAME-A)|---SSRC Aa-->| CNAME-B) |<--SSRC Bv---|(CNAME-B)| 1279 +---------+ +----------+ +---------+ 1281 Figure 10: SRC Using Mixing 1283 8.4. RTP Session Usage by SRS 1285 An SRS that supports recording an audio CS MUST support SRC usage of 1286 separate audio m-lines in SDP, one per CS media direction. An SRS 1287 that supports recording an video CS MUST support SRC usage of 1288 separate video m-lines in SDP, one per CS media direction. 1289 Therefore, for an SRS supporting a typical audio call, the SRS has to 1290 support receiving at least two audio m-lines. For an SRS supporting 1291 a typical audio and video call, the SRS has to support receiving at 1292 least four total m-lines in the SDP, two audio m-lines and two video 1293 m-lines. 1295 These requirements allow an SRS to be implemented that supports video 1296 only, without requiring support for audio recording. They also allow 1297 an SRS to be implemented that supports recording only one direction 1298 of one stream in a CS; for example, an SRS designed to record 1299 security monitoring cameras that only send (not receive) video 1300 without any audio. These requirements were not written to prevent 1301 other modes being implemented and used, such as using a single m-line 1302 and mixing the separate audio streams together. Rather, the 1303 requirements were written to provide a common base mode to implement 1304 for the sake of interoperability. It is important to note that an 1305 SRS implementation supporting the common base may not record all 1306 media streams in a CS if a participant supports more than one m-line 1307 in a video call, such as one for camera and one for presentation. 1308 SRS implementations may support other modes as well, but have to at 1309 least support the ones above such that they interoperate in the 1310 common base mode for basic interoperability. 1312 9. Metadata 1314 9.1. Procedures at the SRC 1316 The SRC MUST deliver metadata to the SRS in a recording session; the 1317 timing of which SRC sends the metadata depends on when the metadata 1318 becomes available. Metadata SHOULD be provided by the SRC in the 1319 initial INVITE request when establishing the recording session, and 1320 subsequent metadata updates can be provided by the SRC in reINVITE 1321 and UPDATE requests ([RFC3311]) and responses in the recording 1322 session. There are cases that metadata is not available in the 1323 initial INVITE request sent by the SRC, for example, when a recording 1324 session is established in the absence of a communication session, and 1325 the SRC would update the recording session with metadata whenever 1326 metadata becomes available. 1328 Certain metadata attributes are contained in the SDP, and others are 1329 contained in a new content type "application/rs-metadata". The 1330 format of the metadata is described as part of the mechanism in 1331 [I-D.ietf-siprec-metadata]. A new "disposition-type" of Content- 1332 Disposition is defined for the purpose of carrying metadata and the 1333 value is "recording-session". The "recording-session" value 1334 indicates that the "application/rs-metadata" content contains 1335 metadata to be handled by the SRS, and the disposition can be carried 1336 in either INVITE or UPDATE requests or responses sent by the SRC. 1338 Metadata sent by the SRC can be categorized as either a full metadata 1339 snapshot or partial update. A full metadata snapshot describes all 1340 the recorded streams and all metadata associated with the recording 1341 session. The SRC MAY send a full metadata snapshot at any time. The 1342 SRC MAY send a partial update if a full metadata snapshot has 1343 previously been sent. If the SRC receives a snapshot request from 1344 the SRS, then it MUST immediately send a full metadata snapshot. 1346 The SRC MAY send metadata (either a full metadata shapshot or a 1347 partial update) in either an INVITE request, an UPDATE request 1348 ([RFC3311]), or in the final (200) response to an offerless INVITE 1349 from the SRS. If any of the metadata being sent contains a reference 1350 to any SDP labels, then the request containing the metadata MUST also 1351 contain an SDP offer that defines those labels. 1353 When an INVITE or UPDATE request contains both an SDP offer and 1354 metadata, then the request body has content type multipart/mixed, 1355 with one subordinate body part containing the SDP offer and another 1356 containing the metadata. When an INVITE contains only an SDP offer 1357 or metadata, then the multipart/mixed container is optional. 1359 The following is an example of a full metadata snapshot sent by the 1360 SRC in the initial INVITE request: 1362 INVITE sip:recorder@example.com SIP/2.0 1363 Via: SIP/2.0/TCP src.example.com;branch=z9hG4bKdf6b622b648d9 1364 From: ;tag=35e195d2-947d-4585-946f-09839247 1365 To: 1366 Call-ID: d253c800-b0d1ea39-4a7dd-3f0e20a 1367 CSeq: 101 INVITE 1368 Max-Forwards: 70 1369 Require: siprec 1370 Accept: application/sdp, application/rs-metadata-request 1371 Contact: ;+sip.src 1372 Content-Type: multipart/mixed;boundary=foobar 1373 Content-Length: [length] 1375 --foobar 1376 Content-Type: application/sdp 1378 v=0 1379 o=SRS 2890844526 2890844526 IN IP4 198.51.100.1 1380 s=- 1381 c=IN IP4 198.51.100.1 1382 t=0 0 1383 m=audio 12240 RTP/AVP 0 4 8 1384 a=sendonly 1385 a=label:1 1387 --foobar 1388 Content-Type: application/rs-metadata 1389 Content-Disposition: recording-session 1391 [metadata content] 1393 Figure 11: Sample INVITE request for the recording session 1395 9.2. Procedures at the SRS 1397 The SRS receives metadata updates from the SRC in INVITE and UPDATE 1398 requests. Since the SRC can send partial updates based on the 1399 previous update, the SRS needs to keep track of the sequence of 1400 updates from the SRC. 1402 In the case of an internal failure at the SRS, the SRS may fail to 1403 recognize a partial update from the SRC. The SRS may be able to 1404 recover from the internal failure by requesting for a full metadata 1405 snapshot from the SRC. Certain errors, such as syntax errors or 1406 semantic errors in the metadata information, are likely caused by an 1407 error on the SRC side, and it is likely the same error will occur 1408 again even when a full metadata snapshot is requested. In order to 1409 avoid repeating the same error, the SRS can simply terminate the 1410 recording session when a syntax error or semantic error is detected 1411 in the metadata. 1413 The SRS MAY explicitly request a full metadata snapshot by sending an 1414 UPDATE request. This request MUST contain a body with a content 1415 disposition type "recording-session", and MUST NOT contain an SDP 1416 body part. Note that the SRS MAY generate an INVITE request without 1417 an SDP offer but this MUST NOT include a metadata snapshot request. 1418 The format of the content is "application/rs-metadata-request", and 1419 the body format is chosen to be a simple text-based format. The 1420 following shows an example: 1422 UPDATE sip:2000@src.exmaple.com SIP/2.0 1423 Via: SIP/2.0/UDP srs.example.com;branch=z9hG4bKdf6b622b648d9 1424 To: ;tag=35e195d2-947d-4585-946f-098392474 1425 From: ;tag=1234567890 1426 Call-ID: d253c800-b0d1ea39-4a7dd-3f0e20a 1427 CSeq: 1 UPDATE 1428 Max-Forwards: 70 1429 Require: siprec 1430 Contact: ;+sip.srs 1431 Accept: application/sdp, application/rs-metadata 1432 Content-Disposition: recording-session 1433 Content-Type: application/rs-metadata-request 1434 Content-Length: [length] 1435 SRS internal error 1437 Figure 12: Metadata Request 1439 The SRS MAY include the reason why a metadata snapshot request is 1440 being made to the SRC in the reason line. This reason line is free 1441 form text, mainly designed for logging purposes on the SRC side. The 1442 processing of the content by the SRC is entirely optional since the 1443 content is for logging only, and the snapshot request itself is 1444 indicated by the use of the application/rs-metadata-request content 1445 type. 1447 When the SRC receives a request for a metadata snapshot, it MUST 1448 immediately provide a full metadata snapshot in a separate INVITE or 1449 UPDATE transaction. Any subsequent partial updates will not be 1450 dependent on any metadata sent prior to this full metadata snapshot. 1452 The metadata received by the SRS can contain ID elements used to 1453 cross reference one element to another. An element containing the 1454 definition of an ID, and an element containing a reference to that ID 1455 will often be received from the same SRC. It is also valid for those 1456 elements to be received from different SRCs, for example, when each 1457 endpoint in the same CS act as an SRC to record the call and a common 1458 ID refers to the same CS. The SRS MUST NOT consider this an error. 1460 9.2.1. Formal Syntax 1462 The formal syntax for the application/rs-metadata-request MIME is 1463 described below using the augmented Backus-Naur Form (BNF) as 1464 described in [RFC5234]. 1466 snapshot-request = srs-reason-line CRLF 1467 srs-reason-line = [TEXT-UTF8-TRIM] 1468 ; TEXT-UTF8-TRIM defined in RFC3261 1470 10. Persistent Recording 1472 Persistent recording is a specific use case outlined in REQ-005 or 1473 Use Case 4 in [RFC6341], where a recording session can be established 1474 in the absence of a communication session. The SRC continuously 1475 records media in a recording session to the SRS even in the absence 1476 of a CS for all user agents that are part of persistent recording. 1477 By allocating recorded streams and continuously sending recorded 1478 media to the SRS, the SRC does not have to prepare new recorded 1479 streams with new SDP offer when a new communication session is 1480 created and also does not impact the timing of the CS. The SRC only 1481 needs to update the metadata when new communication sessions are 1482 created. 1484 When there is no communication sessions running on the devices with 1485 persistent recording, there is no recorded media to stream from the 1486 SRC to the SRS. In certain environments where Network Address 1487 Translator (NAT) is used, typically a minimum of flow activity is 1488 required to maintain the NAT binding for each port opened. Agents 1489 that support Interactive Connectivity Establishment (ICE) solves this 1490 problem. For non-ICE agents, in order not to lose the NAT bindings 1491 for the RTP/RTCP ports opened for the recorded streams, the SRC and 1492 SRS SHOULD follow the recommendations provided in [RFC6263] to 1493 maintain the NAT bindings. 1495 11. IANA Considerations 1497 11.1. Registration of Option Tags 1499 This specification registers two option tags. The required 1500 information for this registration, as specified in [RFC3261], is as 1501 follows. 1503 11.1.1. siprec Option Tag 1505 Name: siprec 1507 Description: This option tag is for identifying the SIP session 1508 for the purpose of recording session only. This is typically not 1509 used in a Supported header. When present in a Require header in a 1510 request, it indicates that the UAS MUST be either a SRC or SRS 1511 capable of handling the contexts of a recording session. 1513 11.1.2. record-aware Option Tag 1515 Name: record-aware 1517 Description: This option tag is to indicate the ability for the 1518 user agent to receive recording indicators in media level or 1519 session level SDP. When present in a Supported header, it 1520 indicates that the UA can receive recording indicators in media 1521 level or session level SDP. 1523 11.2. Registration of media feature tags 1525 This document registers two new media feature tags in the SIP tree 1526 per the process defined in [RFC2506] and [RFC3840] 1528 11.2.1. src feature tag 1530 Media feature tag name: sip.src 1532 ASN.1 Identifier: 25 1534 Summary of the media feature indicated by this tag: This feature 1535 tag indicates that the user agent is a Session Recording Client 1536 for the purpose for Recording Session. 1538 Values appropriate for use with this feature tag: boolean 1540 The feature tag is intended primarily for use in the following 1541 applications, protocols, services, or negotiation mechanisms: This 1542 feature tag is only useful for a Recording Session. 1544 Examples of typical use: Routing the request to a Session 1545 Recording Server. 1547 Security Considerations: Security considerations for this media 1548 feature tag are discussed in Section 11.1 of RFC 3840. 1550 11.2.2. srs feature tag 1552 Media feature tag name: sip.srs 1554 ASN.1 Identifier: 26 1556 Summary of the media feature indicated by this tag: This feature 1557 tag indicates that the user agent is a Session Recording Server 1558 for the purpose for Recording Session. 1560 Values appropriate for use with this feature tag: boolean 1562 The feature tag is intended primarily for use in the following 1563 applications, protocols, services, or negotiation mechanisms: This 1564 feature tag is only useful for a Recording Session. 1566 Examples of typical use: Routing the request to a Session 1567 Recording Client. 1569 Security Considerations: Security considerations for this media 1570 feature tag are discussed in Section 11.1 of RFC 3840. 1572 11.3. New Content-Disposition Parameter Registrations 1574 This document registers a new "disposition-type" value in Content- 1575 Disposition header: recording-session. 1577 recording-session the body describes the metadata information about 1578 the recording session 1580 11.4. Media Type Registration 1582 11.4.1. Registration of MIME Type application/rs-metadata 1584 This document registers the application/rs-metadata MIME media type 1585 in order to describe the recording session metadata. This media type 1586 is defined by the following information: 1588 Media type name: application 1590 Media subtype name: rs-metadata 1592 Required parameters: none 1594 Options parameters: none 1596 11.4.2. Registration of MIME Type application/rs-metadata-request 1598 This document registers the application/rs-metadata-request MIME 1599 media type in order to describe a recording session metadata snapshot 1600 request. This media type is defined by the following information: 1602 Media type name: application 1604 Media subtype name: rs-metadata-request 1606 Required parameters: none 1608 Options parameters: none 1610 11.5. SDP Attributes 1612 This document registers the following new SDP attributes. 1614 11.5.1. 'record' SDP Attribute 1616 Contact names: Leon Portman leon.portman@nice.com, Henry Lum 1617 henry.lum@genesyslab.com 1619 Attribute name: record 1620 Long form attribute name: Recording Indication 1622 Type of attribute: session or media level 1624 Subject to charset: no 1626 This attribute provides the recording indication for the session or 1627 media stream. 1629 Allowed attribute values: on, off, paused 1631 11.5.2. 'recordpref' SDP Attribute 1633 Contact names: Leon Portman leon.portman@nice.com, Henry Lum 1634 henry.lum@genesyslab.com 1636 Attribute name: recordpref 1638 Long form attribute name: Recording Preference 1640 Type of attribute: session or media level 1642 Subject to charset: no 1644 This attribute provides the recording preference for the session or 1645 media stream. 1647 Allowed attribute values: on, off, pause, nopreference 1649 12. Security Considerations 1651 The recording session is fundamentally a standard SIP dialog 1652 [RFC3261], therefore, the recording session can reuse any of the 1653 existing SIP security mechanisms available for securing the session 1654 signaling, the recorded media, and the metadata. The use cases and 1655 requirements document [RFC6341] outlines the general security 1656 considerations, and this document describes specific security 1657 recommendations. 1659 The SRC and SRS MUST support SIP with TLS and MAY support SIPS with 1660 TLS as per [RFC5630]. The Recording Session SHOULD be at least as 1661 secure as the Communication Session, meaning using at least the same 1662 strength of cipher suite as the CS if the CS is secured. For 1663 example, if the CS uses SIPS for signaling and RTP/SAVP for media, 1664 then the RS should not downgrade the level of security in the RS to 1665 SIP or plain RTP since doing so will mean an automatic security 1666 downgrade for the CS. In deployments where the SRC and the SRS are 1667 in the same administrative domain and the same physical switch that 1668 prevents outside user access, some SRCs may choose to lower the level 1669 of security when establishing a recording session. While physically 1670 securing the SRC and SRS may prevent an outside attacker from 1671 accessing important call recordings, this still does not prevent an 1672 inside attacker from accessing the internal network to gain access to 1673 the call recordings. 1675 12.1. Authentication and Authorization 1677 At the transport level, the recording session uses TLS authentication 1678 to validate the authenticity of the SRC and SRS. The SRC and SRS 1679 MUST implement TLS mutual authentication for establishing the 1680 recording session. Whether the SRC/SRS chooses to use TLS mutual 1681 authentication is a deployment decision. In deployments where the 1682 SRC and the SRS are in the same administrative domain, the SRC and 1683 SRS may choose not to authenticate each other, or to have the SRC 1684 authenticate the SRS only, as there is an inherent trust relation 1685 between the SRC and the SRS when they are hosted in the same 1686 administrative domain. In deployments where the SRS can be hosted on 1687 a different administrative domain, it is important to perform mutual 1688 authentication to ensure the authenticity of both the SRC and the SRS 1689 before transmitting any recorded media. The risk of not 1690 authenticating the SRS is that the recording may be sent to a 1691 compromised SRS and that a sensitive call recording will be obtained 1692 by an attacker. On the other hand, the risk of not authenticating 1693 the SRC is that an SRS will accept calls from an unknown SRC and 1694 allow potential forgery of call recordings. 1696 There may be scenarios in which the signaling between the SRC and SRS 1697 is not direct, e.g. a SIP proxy exists between the SRC and the SRS. 1698 In such scenarios, each hop is subject to the TLS mutual 1699 authentication constraint and transitive trust at each hop is 1700 utilized. Additionally, an SRC or SRS may use other existing SIP 1701 mechanisms available, including but not limited to, Digest 1702 Authentication [RFC3261], Asserted Identity [RFC3325], and Connected 1703 Identity [RFC4916]. 1705 The SRS may have its own set of recording policies to authorize 1706 recording requests from the SRC. The use of recording policies is 1707 outside the scope of the Session Recording Protocol. 1709 12.2. RTP handling 1711 In many scenarios it will be critical for the media transported 1712 between the SRC and the SRS to be protected. Media encryption is an 1713 important element in the overall SIPREC solution; therefore the SRC 1714 and the SRS MUST support RTP/SAVP [RFC3711] and RTP/SAVPF [RFC5124]. 1715 RTP/SAVP and RTP/SAVPF provide media encryption, integrity 1716 protection, replay protection, and a limited form of source 1717 authentication. They do not contain or require a specific keying 1718 mechanism. At a minimum, the SRC and SRS MUST support the SDP 1719 Security Descriptions (SDES) key negotiation mechanism [RFC4568]. 1720 For cases in which DTLS-SRTP is used to encrypt a CS media stream, an 1721 SRC may use SRTP Encrypted Key Transport (EKT) 1722 [I-D.ietf-avt-srtp-ekt] in order to use SRTP-SDES in the RS without 1723 needing to re-encrypt the media. 1725 When RTP/SAVP or RTP/SAVPF is used, an SRC can choose to use the same 1726 or different keys in the RS than the ones used in the CS. Some SRCs 1727 are designed to simply replicate RTP packets from a CS media stream 1728 to the SRS, in which case the SRC will use the same key in the RS as 1729 used in the CS. In this case, the SRC MUST secure the SDP containing 1730 the keying material in the RS with at least the same level of 1731 security as in the CS. The risk of lowering the level of security in 1732 the RS is that it will effectively become a downgrade attack on the 1733 CS since the same key is used for both CS and RS. 1735 SRCs that decrypt an encrypted CS media stream and re-encrypt it when 1736 sending it to the SRS MUST use a different key for the RS media 1737 stream than what is used for the CS media stream, to ensure that it 1738 is not possible for someone who has the key for the CS media stream 1739 to access recorded data they are not authorized to access. 1741 12.3. Metadata 1743 Metadata contains sensitive information such as the address of record 1744 of the participants and other extension data placed by the SRC. It 1745 is essential to protect the content of the metadata in the RS. Since 1746 metadata is a content type transmitted in SIP signaling, metadata 1747 SHOULD be protected at the transport level by SIPS/TLS. 1749 12.4. Storage and playback 1751 While storage and playback of the call recording is beyond the scope 1752 of this document, it is worthwhile to mention here that it is also 1753 important for the recording storage and playback to provide a level 1754 of security that is comparable to the communication session. It 1755 would defeat the purpose of securing both the communication session 1756 and the recording session mentioned in the previous sections if the 1757 recording can be easily played back with a simple unsecured HTTP 1758 interface without any form of authentication or authorization. 1760 13. Acknowledgements 1762 We want to thank John Elwell, Paul Kyzivat, Partharsarathi R, Ram 1763 Mohan R, Hadriel Kaplan, Adam Roach, Miguel Garcia, Thomas Stach, 1764 Muthu Perumal, Dan Wing, and Magnus Westerlund for their valuable 1765 comments and inputs to this document. 1767 14. References 1769 14.1. Normative References 1771 [I-D.ietf-siprec-metadata] 1772 R, R., Ravindran, P., and P. Kyzivat, "Session Initiation 1773 Protocol (SIP) Recording Metadata", draft-ietf-siprec- 1774 metadata-12 (work in progress), May 2013. 1776 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1777 Requirement Levels", BCP 14, RFC 2119, March 1997. 1779 [RFC2506] Holtman, K., Mutz, A., and T. Hardie, "Media Feature Tag 1780 Registration Procedure", BCP 31, RFC 2506, March 1999. 1782 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 1783 A., Peterson, J., Sparks, R., Handley, M., and E. 1784 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 1785 June 2002. 1787 [RFC3264] Rosenberg, J. and H. Schulzrinne, "An Offer/Answer Model 1788 with Session Description Protocol (SDP)", RFC 3264, June 1789 2002. 1791 [RFC3840] Rosenberg, J., Schulzrinne, H., and P. Kyzivat, 1792 "Indicating User Agent Capabilities in the Session 1793 Initiation Protocol (SIP)", RFC 3840, August 2004. 1795 [RFC4574] Levin, O. and G. Camarillo, "The Session Description 1796 Protocol (SDP) Label Attribute", RFC 4574, August 2006. 1798 [RFC5234] Crocker, D. and P. Overell, "Augmented BNF for Syntax 1799 Specifications: ABNF", STD 68, RFC 5234, January 2008. 1801 14.2. Informative References 1803 [I-D.ietf-avt-srtp-ekt] 1804 Wing, D., McGrew, D., and K. Fischer, "Encrypted Key 1805 Transport for Secure RTP", draft-ietf-avt-srtp-ekt-03 1806 (work in progress), October 2011. 1808 [I-D.ietf-siprec-architecture] 1809 Hutton, A., Portman, L., Jain, R., and K. Rehor, "An 1810 Architecture for Media Recording using the Session 1811 Initiation Protocol", draft-ietf-siprec-architecture-08 1812 (work in progress), May 2013. 1814 [RFC3311] Rosenberg, J., "The Session Initiation Protocol (SIP) 1815 UPDATE Method", RFC 3311, October 2002. 1817 [RFC3325] Jennings, C., Peterson, J., and M. Watson, "Private 1818 Extensions to the Session Initiation Protocol (SIP) for 1819 Asserted Identity within Trusted Networks", RFC 3325, 1820 November 2002. 1822 [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. 1823 Jacobson, "RTP: A Transport Protocol for Real-Time 1824 Applications", STD 64, RFC 3550, July 2003. 1826 [RFC3551] Schulzrinne, H. and S. Casner, "RTP Profile for Audio and 1827 Video Conferences with Minimal Control", STD 65, RFC 3551, 1828 July 2003. 1830 [RFC3711] Baugher, M., McGrew, D., Naslund, M., Carrara, E., and K. 1831 Norrman, "The Secure Real-time Transport Protocol (SRTP)", 1832 RFC 3711, March 2004. 1834 [RFC4568] Andreasen, F., Baugher, M., and D. Wing, "Session 1835 Description Protocol (SDP) Security Descriptions for Media 1836 Streams", RFC 4568, July 2006. 1838 [RFC4585] Ott, J., Wenger, S., Sato, N., Burmeister, C., and J. Rey, 1839 "Extended RTP Profile for Real-time Transport Control 1840 Protocol (RTCP)-Based Feedback (RTP/AVPF)", RFC 4585, July 1841 2006. 1843 [RFC4916] Elwell, J., "Connected Identity in the Session Initiation 1844 Protocol (SIP)", RFC 4916, June 2007. 1846 [RFC4961] Wing, D., "Symmetric RTP / RTP Control Protocol (RTCP)", 1847 BCP 131, RFC 4961, July 2007. 1849 [RFC5104] Wenger, S., Chandra, U., Westerlund, M., and B. Burman, 1850 "Codec Control Messages in the RTP Audio-Visual Profile 1851 with Feedback (AVPF)", RFC 5104, February 2008. 1853 [RFC5124] Ott, J. and E. Carrara, "Extended Secure RTP Profile for 1854 Real-time Transport Control Protocol (RTCP)-Based Feedback 1855 (RTP/SAVPF)", RFC 5124, February 2008. 1857 [RFC5168] Levin, O., Even, R., and P. Hagendorf, "XML Schema for 1858 Media Control", RFC 5168, March 2008. 1860 [RFC5630] Audet, F., "The Use of the SIPS URI Scheme in the Session 1861 Initiation Protocol (SIP)", RFC 5630, October 2009. 1863 [RFC5761] Perkins, C. and M. Westerlund, "Multiplexing RTP Data and 1864 Control Packets on a Single Port", RFC 5761, April 2010. 1866 [RFC6263] Marjou, X. and A. Sollaud, "Application Mechanism for 1867 Keeping Alive the NAT Mappings Associated with RTP / RTP 1868 Control Protocol (RTCP) Flows", RFC 6263, June 2011. 1870 [RFC6341] Rehor, K., Portman, L., Hutton, A., and R. Jain, "Use 1871 Cases and Requirements for SIP-Based Media Recording 1872 (SIPREC)", RFC 6341, August 2011. 1874 [RFC7022] Begen, A., Perkins, C., Wing, D., and E. Rescorla, 1875 "Guidelines for Choosing RTP Control Protocol (RTCP) 1876 Canonical Names (CNAMEs)", RFC 7022, September 2013. 1878 Authors' Addresses 1880 Leon Portman 1881 NICE Systems 1882 22 Zarhin Street 1883 P.O. Box 690 1884 Ra'anana 4310602 1885 Israel 1887 Email: leon.portman@gmail.com 1889 Henry Lum (editor) 1890 Genesys 1891 1380 Rodick Road, Suite 201 1892 Markham, Ontario L3R4G5 1893 Canada 1895 Email: henry.lum@genesyslab.com 1896 Charles Eckel 1897 Cisco 1898 170 West Tasman Drive 1899 San Jose, CA 95134 1900 United States 1902 Email: eckelcu@cisco.com 1904 Alan Johnston 1905 Avaya 1906 St. Louis, MO 63124 1908 Email: alan.b.johnston@gmail.com 1910 Andrew Hutton 1911 Unify 1912 Brickhill Street 1913 Milton Keynes MK15 0DJ 1914 United Kingdom 1916 Email: andrew.hutton@unify.com