Internet Engineering Task Force MMUSIC WG Internet Draft H. Schulzrinne draft-ietf-mmusic-rfc2326bis-11.txt Columbia U. October 24, 2005 A. Rao Expires: April, 2006 Cisco R. Lanphier RealNetworks Magnus Westerlund Ericsson A. Narasimhan Overture Real Time Streaming Protocol 2.0 (RTSP) STATUS OF THIS MEMO By submitting this Internet-Draft, each author represents that any applicable patent or other IPR claims of which he or she is aware have been or will be disclosed, and any of which he or she becomes aware will be disclosed, in accordance with Section 6 of BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet- Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress". The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html Abstract This memorandum defines RTSP version 2.0 which is a revision of the Proposed Standard RTSP version 1.0 which is defined in RFC 2326. The Real Time Streaming Protocol, or RTSP, is an application-level protocol for control over the delivery of data with real-time properties. RTSP provides an extensible framework to enable controlled, on-demand delivery of real-time data, such as audio and video. Sources of data can include both live data feeds and stored clips. This protocol is intended to control multiple data delivery H. Schulzrinne et. al. [Page 1] Internet Draft RTSP October 24, 2005 sessions, provide a means for choosing delivery channels such as UDP, multicast UDP and TCP, and provide a means for choosing delivery mechanisms based upon RTP (RFC 3550). H. Schulzrinne et. al. [Page 2] Internet Draft RTSP October 24, 2005 H. Schulzrinne et. al. [Page 3] Internet Draft RTSP October 24, 2005 1 Introduction 1.1 RTSP Specification Update This memorandum specifies RTSP 2.0 which is an update of RTSP 1.0, a proposed standard defined in RFC 2326 [24]. The goal of this version is to correct the many flaws that have been identified in RTSP 1.0 since its publication. The corrections are such that full backwards compatibility was impossible. Thus a new version was decided the most appropriate solution to get a more functional protocol. There are no plans to revise RTSP 1.0. Appendix H catalogs the changes of this version in relation to RTSP 1.0. A few open issues still remain to be resolved, and are listed in appendix G. These are intended to be close quickly. A list of bugs against RFC 2326 is available at "http://rtspspec.sourceforge.net". These bugs should be taken into account when reading this memorandum. Input on the unresolved bugs and other issues can be sent via e-mail to the MMUSIC WG's mailing list mmusic@ietf.org and the authors. RTSP 2.0 is reduced in functionality in regards to RTSP 1.0 and aims at specifying the RTSP core, functionality and rules for extensions, and basic interaction with the media delivery protocol RTP. Any other functionality would be need to be published as extension documents. The Working group has discussed a number of different proposals to extensions and is currently working on: o NAT and FW traversal mechanisms for RTSP are described in a document called "How to make Real-Time Streaming Protocol (RTSP) traverse Network Address Translators (NAT) and interact with Firewalls." [25]. 1.2 Purpose The Real-Time Streaming Protocol (RTSP) establishes and controls one or several time-synchronized streams of continuous media such as audio and video. Put simply, RTSP acts as a "network remote control" for multimedia servers. There is no notion of an RTSP connection in the protocol. Instead, an RTSP server maintains a session labeled by an identifier to associate groups of media streams and their states. An RTSP session is not tied to a transport-level connection such as a TCP connection. During a session, a client may open and close many reliable transport connections to the server to issue RTSP requests for that session. H. Schulzrinne et. al. [Page 9] Internet Draft RTSP October 24, 2005 This memorandum describes the use of RTSP over a reliable connection based transport level protocol such as TCP. RTSP may be implemented over an unreliable connectionless transport protocol such as UDP. While nothing in RTSP precludes this, additional definition of this problem area needs to be handled as an extension to the core specification. The mechanisms of RTSP's operation over UDP were left out of this spec. because they were poorly defined in RFC 2326 [24] and the tradeoff in size and complexity of this memorandum for a small gain in a limited problem space was not deemed justifiable. The set of streams to be controlled in an RTSP session is defined by a presentation description. This memorandum does not define a format for the presentation description. However appendix C defines how SDP [1] is used for this purpose. The streams controlled by RTSP may use RTP [2] for their data transport, but the operation of RTSP does not depend on the transport mechanism used to carry continuous media. RTSP is intentionally similar in syntax and operation to HTTP/1.1 [3] so that extension mechanisms to HTTP can in most cases also be added to RTSP. However, RTSP differs in a number of important aspects from HTTP: o RTSP introduces a number of new methods and has a different protocol identifier. o RTSP has the notion of a session built into the protocol. o An RTSP server needs to maintain state by default in almost all cases, as opposed to the stateless nature of HTTP. o Both an RTSP server and client can issue requests. o Data is usually carried out-of-band by a different protocol. Session descriptions returned in a DESCRIBE response (see Section 11.2) and interleaving of RTP with RTSP over TCP are exceptions to this rule (see Section 12). o RTSP is defined to use ISO 10646 (UTF-8) rather than ISO 8859-1, consistent with HTML internationalization efforts [26]. o The Request-URI always contains the absolute URI. Because of backward compatibility with a historical blunder, HTTP/1.1 [3] carries only the absolute path in the request and puts the host name in a separate header field. H. Schulzrinne et. al. [Page 10] Internet Draft RTSP October 24, 2005 This makes "virtual hosting" easier, where a single host with one IP address hosts several document trees. The protocol supports the following operations: Retrieval of media from media server: The client can either request a presentation description via RTSP DESCRIBE, HTTP or some other method. If the presentation is being multicast, the presentation description contains the multicast addresses and ports to be used for the continuous media. If the presentation is to be sent only to the client via unicast, the client provides the destination of necessity. Invitation of a media server to a conference: A media server can be "invited" to join an existing conference to play back media into the presentation. This mode is useful for example distributed teaching applications. Several parties in the conference may take turns "pushing the remote control buttons". RTSP requests may be handled by proxies, tunnels and caches as in HTTP/1.1 [3]. 1.3 Notational Conventions Since many of the definitions and syntax are identical to HTTP/1.1, this specification only points to the section where they are defined rather than copying it. For brevity, [HX.Y] is to be taken to refer to Section X.Y of the current HTTP/1.1 specification (RFC 2616 [3]). All the mechanisms specified in this document are described in both prose and the Augmented Backus-Naur form (ABNF) described in detail in RFC 4234 [4]. Indented and smaller-type paragraphs are used to provide informative background and motivation. This is intended to give readers who were not involved with the formulation of the specification an understanding of why things are the way they are in RTSP. The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [5]. The word, "unspecified" is used to indicate functionality or features that are not defined in this specification. Such functionality cannot be used in a standardized manner without further definition in an extension specification to RTSP. H. Schulzrinne et. al. [Page 11] Internet Draft RTSP October 24, 2005 1.4 Terminology Some of the terminology has been adopted from HTTP/1.1 [3]. Terms not listed here are defined as in HTTP/1.1. Aggregate control: The concept of controlling multiple streams using a single timeline, generally maintained by the server. A client, for example, uses aggregate control when it issues a single play or pause message to simultaneously control both the audio and video in a movie. Aggregate control URI: The URI used in an RTSP request to refer to and control an aggregated session. It normally, but not always, corresponds to the presentation URI specified in the session description. See Section 11.3 for more information. Conference: a multiparty, multimedia presentation, where "multi" implies greater than or equal to one. Client: The client requests media service from the media server. Connection: A transport layer virtual circuit established between two programs for the purpose of communication. Container file: A file which may contain multiple media streams which often constitutes a presentation when played together. The concept of a container file is not embedded in the protocol. However, RTSP servers may offer aggregate control on the media streams within these files. Continuous media: Data where there is a timing relationship between source and sink; that is, the sink needs to reproduce the timing relationship that existed at the source. The most common examples of continuous media are audio and motion video. Continuous media can be real-time (interactive or conversational), where there is a "tight" timing relationship between source and sink, or streaming (playback), where the relationship is less strict. Entity: The information transferred as the payload of a request or response. An entity consists of meta-information in the form of entity-header fields and content in the form of an entity-body, as described in Section 8. Feature-tag: A tag representing a certain set of functionality, i.e. a feature. H. Schulzrinne et. al. [Page 12] Internet Draft RTSP October 24, 2005 Live: Normally used to describe a presentation or session with media coming from an ongoing event. This generally results in that the session has a unbound or only loosely defined duration, and that no seek operations are possible. Media initialization: Datatype/codec specific initialization. This includes such things as clock rates, color tables, etc. Any transport-independent information which is required by a client for playback of a media stream occurs in the media initialization phase of stream setup. Media parameter: Parameter specific to a media type that may be changed before or during stream playback. Media server: The server providing playback services for one or more media streams. Different media streams within a presentation may originate from different media servers. A media server may reside on the same host or on a different host from which the presentation is invoked. Media server indirection: Redirection of a media client to a different media server. (Media) stream: A single media instance, e.g., an audio stream or a video stream as well as a single whiteboard or shared application group. When using RTP, a stream consists of all RTP and RTCP packets created by a source within an RTP session. Message: The basic unit of RTSP communication, consisting of a structured sequence of octets matching the syntax defined in Section 19 and transmitted over a connection or a connectionless transport. Non-Aggregated Control: Control of a single media stream. Only possible in RTSP sessions with a single media. Participant: Member of a conference. A participant may be a machine, e.g., a playback server. Presentation: A set of one or more streams presented to the client as a complete media feed and described by a presentation description as defined below. Presentations with more than one media stream is often handled in RTSP under aggregate control. Presentation description: A presentation description contains information about one or more media streams within a H. Schulzrinne et. al. [Page 13] Internet Draft RTSP October 24, 2005 presentation, such as the set of encodings, network addresses and information about the content. Other IETF protocols such as SDP (RFC 2327 [1]) use the term "session" for a presentation. The presentation description may take several different formats, including but not limited to the session description protocol format, SDP. Response: An RTSP response. If an HTTP response is meant, that is indicated explicitly. Request: An RTSP request. If an HTTP request is meant, that is indicated explicitly. Request-URI: The URI used in a request to indicate the resource on which the request is to be performed. RTSP agent: Refers to either an RTSP client, an RTSP server, or an RTSP Proxy. In this specification, there are many capabilities that are common to these three entities such as the capability to send requests or receive responses. This term will be used when describing functionality that is applicable to all three of these entities. RTSP session: A stateful abstraction upon which the main control methods of RTSP operate. An RTSP session is a server entity; it is created, maintained and destroyed by the server. It is established by an RTSP server upon the completion of a successful SETUP request (when 200 OK response is sent) and is labelled by a session identifier at that time. The session exists until timed out by the server or explicitly removed by a TEARDOWN request. An RTSP session is a stateful entity; an RTSP server maintains an explicit session state machine (see Appendix A) where most state transitions are triggered by client requests. The existence of a session implies the existence of state about the session's media streams and their respective transport mechanisms. A given session can have zero or more media streams associated with it. An RTSP server uses the session to aggregate control over multiple media streams. Transport initialization: The negotiation of transport information (e.g., port numbers, transport protocols) between the client and the server. URI: Universal Resource Identifier, see RFC 3986 [6]. In RTSP the used URIs are as general rule in fact URL's as they gives an location for the resource. As URLs are a subset of URIs, they will be referred to as URIs to cover also the H. Schulzrinne et. al. [Page 14] Internet Draft RTSP October 24, 2005 cases when an RTSP URI would not be an URL. URL: Universal Resource Locator, is an URI which identifies the resource through its primary access mechanism, rather than identifying the resource by name or by some other attribute(s) of that resource. 1.5 Protocol Properties RTSP has the following properties: Extendable: New methods and parameters can be easily added to RTSP. Easy to parse: RTSP can be parsed by standard HTTP or MIME parsers. Secure: RTSP re-uses web security mechanisms, either at the transport level (TLS, RFC 2246 [7]) or within the protocol itself. All HTTP authentication mechanisms such as basic (RFC 2616 [3]) and digest authentication (RFC 2617 [8]) are directly applicable. Transport-independent: RTSP does not preclude the use of an unreliable datagram protocol (UDP) (RFC 768 [9]) as it would be possible to implement application-level reliability. The use of a connectionless datagram protocol such as UDP requires additional definition that may be provided as extensions to the core RTSP specification. The usage of the reliable stream protocol TCP (RFC 793 [10]) and secured reliable stream protocol TLS over TCP [7] is what is currently defined as transport protocol of RTSP messages. Multi-server capable: Each media stream within a presentation can reside on a different server. The client automatically establishes several concurrent control sessions with the different media servers. Media synchronization is performed at the transport level. Separation of stream control and conference initiation: Stream control is divorced from inviting a media server to a conference. In particular, SIP [27] or H.323 [28] may be used to invite a server to a conference. Suitable for professional applications: RTSP supports frame- level accuracy through SMPTE time stamps to allow remote digital editing. H. Schulzrinne et. al. [Page 15] Internet Draft RTSP October 24, 2005 Presentation description neutral: The protocol does not impose a particular presentation description or metafile format and can convey the type of format to be used. However, the presentation description is required to contain at least one RTSP URI. Proxy and firewall friendly: The protocol should be readily handled by both application and transport-layer (SOCKS [29]) firewalls. A firewall may need to understand the SETUP method to open a "hole" for the media stream. HTTP-friendly: Where sensible, RTSP reuses HTTP concepts, so that the existing infrastructure can be reused. This infrastructure includes PICS (Platform for Internet Content Selection [30,31]) for associating labels with content. However, RTSP does not just add methods to HTTP since the controlling continuous media requires server state in most cases. Appropriate server control: If a client can start a stream, it needs to be able to stop a stream. Servers should not start streaming to clients in such a way that clients cannot stop the stream. Transport negotiation: The client can negotiate the transport method prior to actually needing to process a continuous media stream. 1.6 Extending RTSP Since not all media servers have the same functionality, media servers by necessity will support different sets of requests. For example: o A server may not be capable of seeking (absolute positioning) if it is to support live events only. o Some servers may not support setting stream parameters and thus not support GET_PARAMETER and SET_PARAMETER. o Some server may support an RTSP extension. A server SHOULD implement all header fields described in Section 14. It is up to the creators of presentation descriptions not to ask the impossible of a server. This situation is similar in HTTP/1.1 [3], where the methods described in [H19.5] are not likely to be supported across all servers. H. Schulzrinne et. al. [Page 16] Internet Draft RTSP October 24, 2005 RTSP can be extended in three ways, listed here in order of the magnitude of changes supported: o Existing methods can be extended with new parameters, e.g. headers, as long as these parameters can be safely ignored by the recipient. If the client needs negative acknowledgement when a method extension is not supported, a tag corresponding to the extension may be added in the Require: field (see Section 14.37). o New methods can be added. If the recipient of the message does not understand the request, it responds with error code 501 (Not Implemented) and the sender should not attempt to use this method again. A client may also use the OPTIONS method to inquire about methods supported by the server. The server MUST list the methods it supports using the Public response header. o A new version of the protocol can be defined, allowing almost all aspects (except the position of the protocol version number) to change. The basic capability discovery mechanism can be used to both discover support for a certain feature and to ensure that a feature is available when performing a request. For detailed explanation of this see section 10. 1.7 Overall Operation Each presentation and media stream is identified by an RTSP URI. The overall presentation and the properties of the media the presentation is made up of are defined by a presentation description file, the format of which is outside the scope of this specification. The presentation description file may be obtained by the client using HTTP or other means such as email and may not necessarily be stored on the media server. For the purposes of this specification, a presentation description is assumed to describe one or more presentations, each of which maintains a common time axis. For simplicity of exposition and without loss of generality, it is assumed that the presentation description contains exactly one such presentation. A presentation may contain several media streams. The presentation description file contains a description of the media streams making up the presentation, including their encodings, language, and other parameters that enable the client to choose the most appropriate combination of media. In this presentation description, each media stream that is individually controllable by H. Schulzrinne et. al. [Page 17] Internet Draft RTSP October 24, 2005 RTSP is identified by an RTSP URI, which points to the media server handling that particular media stream and names the stream stored on that server. Several media streams can be located on different servers; for example, audio and video streams can be split across servers for load sharing. The description also enumerates which transport methods the server is capable of. Besides the media parameters, the network destination address and port need to be determined. Several modes of operation can be distinguished: Unicast: The media is transmitted to the source of the RTSP request, with the port number chosen by the client. Alternatively, the media is transmitted on the same reliable stream as RTSP. Multicast, server chooses address: The media server picks the multicast address and port. This is the typical case for a live or near-media-on-demand transmission. Multicast, client chooses address: If the server is to participate in an existing multicast conference, the multicast address, port and encryption key are given by the conference description, established by means outside the scope of this specification, for example by a SIP created conference. 1.8 RTSP States RTSP controls a stream which may be sent via a separate protocol, independent of the control channel. For example, RTSP control may be transported on a TCP connection while the media data is conveyed via UDP. Thus, data delivery continues even if no RTSP requests are received by the media server. Also, during its lifetime, a single media stream may be controlled by RTSP requests issued sequentially on different TCP connections. Therefore, the server needs to maintain "session state" to be able to correlate RTSP requests with a stream. The state transitions are described in Appendix A. Many methods in RTSP do not contribute to state. However, the following play a central role in defining the allocation and usage of stream resources on the server: SETUP, PLAY, PAUSE, REDIRECT, and TEARDOWN. SETUP: Causes the server to allocate resources for a stream and create an RTSP session. PLAY: Starts data transmission on a stream allocated via SETUP. H. Schulzrinne et. al. [Page 18] Internet Draft RTSP October 24, 2005 PAUSE: Temporarily halts a stream without freeing server resources. REDIRECT: Indicates that the session should be moved to new server / location TEARDOWN: Frees resources associated with the stream. The RTSP session ceases to exist on the server. RTSP methods that contribute to state use the Session header field (Section 14.42) to identify the RTSP session whose state is being manipulated. The server generates session identifiers in response to SETUP requests (Section 11.3). 1.9 Relationship with Other Protocols RTSP has some overlap in functionality with HTTP. It also may interact with HTTP in that the initial contact with streaming content is often to be made through a web page. The current protocol specification aims to allow different hand-off points between a web server and the media server implementing RTSP. For example, the presentation description can be retrieved using HTTP or RTSP, which reduces round trips in web-browser-based scenarios, yet also allows for stand alone RTSP servers and clients which do not rely on HTTP at all. However, RTSP differs fundamentally from HTTP in that most data delivery takes place out-of-band in a different protocol. HTTP is an asymmetric protocol where the client issues requests and the server responds. In RTSP, both the media client and media server can issue requests. RTSP requests are also stateful; they may set parameters and continue to control a media stream long after the request has been acknowledged. Re-using HTTP functionality has advantages in at least two areas, namely security and proxies. The requirements are very similar, so having the ability to adopt HTTP work on caches, proxies and authentication is valuable. RTSP assumes the existence of a presentation description format that can express both static and temporal properties of a presentation containing several media streams. Session Description Protocol (SDP) [1] is generally the format of choice; however, RTSP is not bound to it. For data delivery, most real-time media will use RTP as a transport protocol. While RTSP works well with RTP, it is not tied to RTP. 2 RTSP Use Cases H. Schulzrinne et. al. [Page 19] Internet Draft RTSP October 24, 2005 This section describes the most important and considered use cases for RTSP. They are listed in descending order of importance in regards to ensuring that all necessary functionality is present. This specification does only fully support usage of the two first. Also in these first two cases, there are special cases or exceptions that are not supported without extensions, e.g. the redirection of media to another address than the controlling entity. 2.1 On-demand Playback of Stored Content An RTSP capable server stores content suitable for being streamed to a client. A client desiring playback of any of the stored content then uses RTSP to set up and configure the media transport required for the desired content. Then RTSP is used to initiate, halt and manipulate the actual transmission (playout) of the content. There are also requirement on being able to use RTSP to carry necessary description and synchronization information for the content. The above high level description can be broken down into a number of functionalities that RTSP needs to be capable of. Presentation Description: The possibility to carry initialization information about the presentation (content), for example, which media codec(s) that are needed for the content. Other information that are important; how many media stream that the presentation contains; what transport protocols used for the media streams; and identifiers for these media streams. This information is required before setup of the content is possible. The information is also needed by the client to determine if it is capable at all to support the content. This information is not required to be sent using RTSP, instead other external protocols can be utilized to transport presentation descriptions. Two good examples are the use of HTTP [3] or email to fetch or receive presentation descriptions like SDP [1]. .XP Setup: Performing setup of some or all of the media streams in a presentation. The setup itself consist of determining which protocols for media transport to use; the necessary parameters for the protocol, like addresses and ports. .XP Control of Transmission: After the necessary media streams has been established the client can request the server to start transmitting the content. There is need to allow the client to at arbitrary times start or stop the transmission of the content. There are also exist need to be able to start the transmission at an any point in the timeline of the presentation. .XP Synchronization: For media transport protocols like RTP [16] it might be beneficial to carry synchronization information within RTSP. Either due to the H. Schulzrinne et. al. [Page 20] Internet Draft RTSP October 24, 2005 lack of inter media synchronization within the protocol itself, or the potential delay before the synchronization is established (which is the case for RTP when using RTCP). .XP Termination There is also need to be able to terminate the established contexts. For this use cases there is a number of assumption about how it works. These are listed below: On-Demand content: The content available is stored at the server and can be accessed at any time during a time period when it is intended to be available. .XP Independent sessions: A server is capable of serving a number of clients simultaneously, including from the same piece of content at different points in that presentations time-line. .XP Unicast Transport: Content for each individual client is transmitted to them using unicast traffic. It is also possible to redirect the media traffic to another destination than where the entity controlling traffic uses. However allowing this without appropriate mechanisms for checking that the destination approves of this is allows for distributed denial of service attacks (DDoS). 2.2 Unicast distribution of Live Content This use cases is not that different from the above on-demand content case (see section 2.1. The difference is really the restriction the content itself establish. Live content is continuously distributed as it becomes available from a source, i.e. the main difference to on- demand is that one starts distributing content before the end of it has become available to the server. In many cases the consumer of live content is only interested in consuming what is actually happens "now", i.e. very similar to broadcast TV. However in this case it is assumed that there exist no broadcast or multicast channel to the users, and instead the server functions as a distribution node, sending the same content to multiple receivers, using unicast traffic between server and client. This unicast traffic and the transport parameters are individually negotiated for each receiving client. Another aspect of live content is that it has often very limited time of availability, as it is only is available for the duration of the event the content covers. A example of such a live content could for example be a music concert, which lasts 2 hour and starts at a predetermined time. Thus there is need to announce when and for how long the live content is available. 2.3 On-demand Playback using Multicast It is possible to use RTSP to request that media is delivered to a multicast group. The entity setting up the session (the controller) H. Schulzrinne et. al. [Page 21] Internet Draft RTSP October 24, 2005 will then control when and what media that is delivered to the group. Also this use case has some potential for denial of service attacks, in this case flooding any multicast group. Therefore there is need for a mechanism indicating that the group actually accepts the traffic from the RTSP server. An open issue in this use case is how one ensures that all receivers listening to the multicast or broadcast receives the session presentation configuring the receivers. 2.4 Inviting a RTSP server into a conference If one has an established conference or group session, it is possible to have a RTSP server distribute media to the whole group. The transmission to the group is simplest controlled by a single participant or leader of the conference. Shared control might be possible, but would require further investigation and possibly extensions. There are some protocol mechanisms missing for this scenario. For reasonable complexity in the media transmission stage, this use case assumes that there exist either multicast or a conference focus that redistribute media to all participants. In some more detail, this use case is intended to be able to handle the following scenario: A conference leader or participant (from here called the controller) has some pre-stored content on a RTSP server that he likes to share with the group. The controller sets up an RTSP session at the streaming server for the content the controller likes to share. The session description for the content is retrieved by the controller. The media destination for the media content is sent to the shared multicast group or conference focus. When desired by the controller, he/she can start and stop the transmission of the media to the conference group. There are several issues with this use case that is not solved by this core specification for RTSP: o Denial of service threat, to avoid a RTSP server from being a unknowing participant of a denial of service attack the server needs to be able to verify the destinations acceptance for the media. Such a mechanism does not yet exist that can be used to verify the approval to received media, instead only policies can be used, which can be made to work in controlled environments. .IP o 2 The problem of distributing the presentation description to all participants in the group. To enable a media receiver to decode the content correctly the media configuration information will need to be distributed reliable to all participants. This will most likely require support from an external protocol. .IP o 2 Passing the control. If it is desired to be able to pass the control of the RTSP session between the participants some support will be required by an external protocol for the necessary exchange of state information and possibly floor control of who is H. Schulzrinne et. al. [Page 22] Internet Draft RTSP October 24, 2005 controlling the RTSP session. So if there interest in this use case further work on the necessary extensions has to be performed. 2.5 Live Content using Multicast This use case does in its simplest form do not require any use of RTSP at all. This is what multicast conferences being announced with SAP and SDP are intended to handle. However in use cases where more advance features like access control to the multicast session is desired, RTSP could be used for session establishment. A client desiring to join a live multicasted media session with cryptographic (encryption) access control could use RTSP in the following way. The source of the session, announces the session and gives all interested to join, a RTSP URI. The client connects to the server and requests the presentation description allowing for configuration the reception. In this step it is possible to use secured transport for the client, and also desired levels of authentication, for example for charging purposes or simply access control. An RTSP link also allows for load balancing between multiple servers. However if this was the only thing that occurred it could probably be solved as simply using HTTP. However for session where the sender likes to keep track of each individual receiver during the session, and possibly use this side channel for pushing out key-updates or other side information that is desirable to be done on a per receiver basis, and the receivers are not know prior to the session start, the state establishment that RTSP provides can be beneficial. In this case a client would establish a RTSP session to the multicast group. The RTSP server will not transmit any media, instead it will simply point to the multicast group. However the client and server will be able to keep the session alive for as long as the receiver participates in the session. Thus enabling, for example server to client pushes of updates. This use cases will most likely not be able to actually implement without some extensions in relation to the server to client push mechanism. Here a method like ANNOUNCE (see RFC 2326 [24] might be suitable, however it will require a RTSP extension to revive the method. 3 Protocol Parameters 3.1 RTSP Version HTTP Specification Section [H3.1] applies, with HTTP replaced by RTSP. This specification defines version 2.0 of RTSP. 3.2 RTSP URI H. Schulzrinne et. al. [Page 23] Internet Draft RTSP October 24, 2005 The "rtsp", "rtsps" schemes are used to refer to network resources via the RTSP protocol. This section defines the scheme-specific syntax and semantics for RTSP URIs. The RTSP URI is case sensitive. An URI scheme "rtspu" was defined in RFC 2326 for transport of RTSP messages over unreliable transport (UDP) and is currently deprecated and reserved, and MUST NOT be used . See Appendix E for further information. Informative RTSP URI syntax: rtsp[u|s]://host[:port]/abspath[?query]#fragment See section 19.2.1 for the formal definition of the RTSP URI syntax. The fragment identifier is used as defined in section 4.1 of [6], i.e. the fragment is to be stripped from the URI by the requestor and not included in the request. The user agent also needs to interpret the value of the fragment based on the media type the request relates to, i.e. the media type indicated in Content-Type header in the response to DESCRIBE. The syntax of any URI query string is unspecified and responder (usually the server) specific. As it is from the requestor an opaque string, it needs to be handled as such. The URI scheme rtsp requires that commands are issued via a reliable protocol (within the Internet, TCP), while the scheme rtsps identifies a reliable transport using secure transport (TLS [7]). If the no port number is provided in the URI, port number 554 SHALL be used. The semantics are that the identified resource can be controlled by RTSP at the server listening for TCP (scheme "rtsp") connections on that port of host, and the Request-URI for the resource is rtsp_URI. For the scheme rtsps the TCP and UDP port 322 is registered and SHALL be assumed. The use of IP addresses in URIs SHOULD be avoided whenever possible (see RFC 1924 [11]). This specification updates the RTSP URI scheme to allow for literal IPv6 addresses using the host specification in RFC 2732 [12]. Note: Using qualified domain names in any URI is one requirement for making it possible for RTSP 1.0 (RFC 2326) implementations of RTSP to use IPv6. A presentation or a stream is identified by a textual media H. Schulzrinne et. al. [Page 24] Internet Draft RTSP October 24, 2005 identifier, using the character set and escape conventions [H3.2] of URIs (RFC 3986 [6]). URIs may refer to a stream or an aggregate of streams, i.e., a presentation. Accordingly, requests described in Section 11 can apply to either the whole presentation or an individual stream within the presentation. Note that some request methods can only be applied to streams, not presentations and vice versa. For example, the RTSP URI: rtsp://media.example.com:554/twister/audiotrack may identify the audio stream within the presentation "twister", which can be controlled via RTSP requests issued over a TCP connection to port 554 of host media.example.com Also, the RTSP URI: rtsp://media.example.com:554/twister identifies the presentation "twister", which may be composed of audio and video streams. This does not imply a standard way to reference streams in URIs. The presentation description defines the hierarchical relationships in the presentation and the URIs for the individual streams. A presentation description may name a stream "a.mov" and the whole presentation "b.mov". The path components of the RTSP URI are opaque to the client and do not imply any particular file system structure for the server. This decoupling also allows presentation descriptions to be used with non-RTSP media control protocols simply by replacing the scheme in the URI. 3.3 Session Identifiers Session identifiers are strings of any arbitrary length. A session identifier MUST be chosen randomly and MUST be at least eight characters long to make guessing it more difficult. (See Section 20.) 3.4 SMPTE Relative Timestamps H. Schulzrinne et. al. [Page 25] Internet Draft RTSP October 24, 2005 A SMPTE relative timestamp expresses time relative to the start of the clip. Relative timestamps are expressed as SMPTE time codes for frame-level access accuracy. The time code has the format hours:minutes:seconds:frames.subframes, with the origin at the start of the clip. The default smpte format is "SMPTE 30 drop" format, with frame rate is 29.97 frames per second. Other SMPTE codes MAY be supported (such as "SMPTE 25") through the use of alternative use of "smpte time". For the "frames" field in the time value can assume the values 0 through 29. The difference between 30 and 29.97 frames per second is handled by dropping the first two frame indices (values 00 and 01) of every minute, except every tenth minute. If the frame value is zero, it may be omitted. Subframes are measured in one-hundredth of a frame. Examples: smpte=10:12:33:20- smpte=10:07:33- smpte=10:07:00-10:07:33:05.01 smpte-25=10:07:00-10:07:33:05.01 3.5 Normal Play Time Normal play time (NPT) indicates the stream absolute position relative to the beginning of the presentation, not to be confused with the Network Time Protocol (NTP) [32]. The timestamp consists of a decimal fraction. The part left of the decimal may be expressed in either seconds or hours, minutes, and seconds. The part right of the decimal point measures fractions of a second. The beginning of a presentation corresponds to 0.0 seconds. Negative values are not defined. The special constant now is defined as the current instant of a live type event. It MAY only be used for live type events, and SHALL NOT be used for on-demand content. NPT is defined as in DSM-CC [33]: "Intuitively, NPT is the clock the viewer associates with a program. It is often digitally displayed on a VCR. NPT advances normally when in normal play mode (scale = 1), advances at a faster rate when in fast scan forward (high positive scale ratio), decrements when in scan reverse (high negative scale ratio) and is fixed in pause mode. NPT is (logically) equivalent to SMPTE time codes." Examples: npt=123.45-125 H. Schulzrinne et. al. [Page 26] Internet Draft RTSP October 24, 2005 npt=12:05:35.3- npt=now- The syntax conforms to ISO 8601 [34]. The npt-sec notation is optimized for automatic generation, the ntp-hhmmss notation for consumption by human readers. The "now" constant allows clients to request to receive the live feed rather than the stored or time-delayed version. This is needed since neither absolute time nor zero time are appropriate for this case. 3.6 Absolute Time Absolute time is expressed as ISO 8601 [34] timestamps, using UTC (GMT). Fractions of a second may be indicated. Example for November 8, 1996 at 14h37 and 20 and a quarter seconds UTC: 19961108T143720.25Z 3.7 Feature-tags Feature-tags are unique identifiers used to designate features in RTSP. These tags are used in Require (Section 14.37), Proxy-Require (Section 14.31), Proxy-Supported (Section 14.32), Unsupported (Section 14.46), and Supported (Section 14.43) header fields. Feature tag needs to indicate which combination of clients, servers, or proxies they applies too. The creator of a new RTSP feature-tag should either prefix the feature-tag with a reverse domain name (e.g., "com.example.mynewfeature" is an apt name for a feature whose inventor can be reached at "example.com"), or register the new feature-tag with the Internet Assigned Numbers Authority (IANA), see IANA Section 21. The usage of feature tags are further described in section 10 that deals with capability handling. 3.8 Entity Tags H. Schulzrinne et. al. [Page 27] Internet Draft RTSP October 24, 2005 Entity tags are opaque strings that are used to compare two entities from the same resource, for example in caches or to optimize setup after a redirect. Further explanation is present in [H3.11]. For explanation on how to compare Entity tags see [H13.3]. Entity tags can be carried in the ETag header (see section 14.21) or in SDP (see section C.1.8). Entity tags are used in RTSP to make some methods conditional. The methods are made conditional through the inclusion of headers, see 14.25 and 14.27. 4 RTSP Message RTSP is a text-based protocol and uses the ISO 10646 character set in UTF-8 encoding (RFC 2279 [13]). Lines SHALL be terminated by CRLF. Text-based protocols make it easier to add optional parameters in a self-describing manner. Since the number of parameters and the frequency of commands is low, processing efficiency is not a concern. Text-based protocols, if done carefully, also allow easy implementation of research prototypes in scripting languages such as Tcl, Visual Basic and Perl. The 10646 character set avoids tricky character set switching, but is invisible to the application as long as US-ASCII is being used. This is also the encoding used for RTCP. ISO 8859-1 translates directly into Unicode with a high-order octet of zero. ISO 8859-1 characters with the most-significant bit set are represented as 1100001x 10xxxxxx. (See RFC 2279 [13]) Requests contain methods, the object the method is operating upon and parameters to further describe the method. Methods are idempotent, unless otherwise noted. Methods are also designed to require little or no state maintenance at the media server. 4.1 Message Types See [H4.1]. 4.2 Message Headers See [H4.2]. 4.3 Message Body See [H4.3] H. Schulzrinne et. al. [Page 28] Internet Draft RTSP October 24, 2005 4.4 Message Length When a message body is included with a message, the length of that body is determined by one of the following (in order of precedence): 1. Any response message which MUST NOT include a message body (such as the 1xx, 204, and 304 responses) is always terminated by the first empty line after the header fields, regardless of the entity-header fields present in the message. (Note: An empty line consists of only CRLF.) 2. If a Content-Length header field (section 14.16) is present, its value in bytes represents the length of the message-body. If this header field is not present, a value of zero is assumed. Unlike an HTTP message, an RTSP message MUST contain a Content-Length header field whenever it contains a message body. Note that RTSP does not (at present) support the HTTP/1.1 "chunked" transfer coding(see [H3.6.1]). Given the moderate length of presentation descriptions returned, the server should always be able to determine its length, even if it is generated dynamically, making the chunked transfer encoding unnecessary. 5 General Header Fields See [H4.5], except that Pragma, Trailer, Transfer-Encoding, Upgrade, and Warning headers are not defined. RTSP further defines the CSeq, and Timestamp. The general headers are listed in table 1: Header Name Comment _________________________________ Cache-Control See section 14.10 Connection See section 14.11 CSeq See section 14.19 Date See section 14.20 Supported See section 14.43 Timestamp See section 14.44 Via See section 14.49 Table 1: The General headers used in RTSP. H. Schulzrinne et. al. [Page 29] Internet Draft RTSP October 24, 2005 6 Request A request messages uses the format outlined below, regardless of the direction of a request, client to server or server to client: o Request line, containing the method to be applied to the resource, the identifier of the resource, and the protocol version in use; o zero or more Header lines, that can be of the following types: general (Section 5), request (Section 6.2), or entity (Section 8.1); o One empty line (CR/LF) to indicate the end of the header section; o Optionally a message body (entity), consisting of one or more lines. the length of the message body in number of bytes is indicated by the Content-Length entity header. 6.1 Request Line The request line provides the key information about the request: What method, on what resources and using which RTSP version. The methods that are defined by this specification are listed in Table 2. Method Defined In Section _________________________________ DESCRIBE Section 11.2 GET_PARAMETER Section 11.7 OPTIONS Section 11.1 PAUSE Section 11.5 PLAY Section 11.4 REDIRECT Section 11.9 SETUP Section 11.3 SET_PARAMETER Section 11.8 TEARDOWN Section 11.6 Table 2: The RTSP Methods The syntax of the RTSP request line is the following: SP SP CRLF H. Schulzrinne et. al. [Page 30] Internet Draft RTSP October 24, 2005 Note: This syntax cannot be freely changed in future versions of RTSP. This line needs to remain parsable by older RTSP implementations since it indicates the RTSP version of the message. In contrast to HTTP/1.1 [3], RTSP requests identify the resource through an absolute RTSP URI (scheme, host, and port)(see section 3.2) rather than just the absolute path. HTTP/1.1 requires servers to understand the absolute URI, but clients are supposed to use the Host request header. This is purely needed for backward-compatibility with HTTP/1.0 servers, a consideration that does not apply to RTSP. An asterisk "*" can be used in the Request-URI to indicate that the request does not apply to a particular resource, but to the server or proxy itself, and is only allowed when the request method does not necessarily apply to a resource. For example: OPTIONS * RTSP/2.0 An OPTIONS in this form will determine the capabilities of the server or the proxy that first receives the request. If the capability of the specific server needs to be determined, without regard to the capability of an intervening proxy, the server should be addressed explicitly with an absolute URI that contains the server's address. For example: OPTIONS rtsp://example.com RTSP/2.0 6.2 Request Header Fields The RTSP headers in Table 3 can be included in a request, as request headers, to modify the specifics of the request. These headers may also be used in the response to a request, as response headers, to modify the specifics of a response (Section 7.1.2). Detailed headers definition are provided in Section 14. H. Schulzrinne et. al. [Page 31] Internet Draft RTSP October 24, 2005 Header Defined in Section _____________________________________ Accept Section 14.1 Accept-Encoding Section 14.3 Accept-Language Section 14.4 Authorization Section 14.7 Bandwidth Section 14.8 Blocksize Section 14.9 From Section 14.23 If-Match Section 14.25 If-Modified-Since Section 14.26 If-None-Match Section 14.27 Proxy-Require Section 14.31 Range Section 14.34 Referer Section 14.35 Require Section 14.37 Scale Section 14.39 Session Section 14.42 Speed Section 14.40 Supported Section 14.43 Transport Section 14.45 User-Agent Section 14.47 Table 3: The RTSP request headers 7 Response [H6] applies except that HTTP-Version is replaced by RTSP-Version. Also, RTSP defines additional status codes and does not define some HTTP codes. The valid response codes and the methods they can be used with are defined in Table 4. After receiving and interpreting a request message, the recipient responds with an RTSP response message. 7.1 Status-Line The first line of a Response message is the Status-Line, consisting of the protocol version followed by a numeric status code, and the textual phrase associated with the status code, with each element separated by SP characters. No CR or LF is allowed except in the final CRLF sequence. SP SP CRLF H. Schulzrinne et. al. [Page 32] Internet Draft RTSP October 24, 2005 7.1.1 Status Code and Reason Phrase The Status-Code element is a 3-digit integer result code of the attempt to understand and satisfy the request. These codes are fully defined in Section 13. The Reason-Phrase is intended to give a short textual description of the Status-Code. The Status-Code is intended for use by automata and the Reason-Phrase is intended for the human user. The client is not required to examine or display the Reason- Phrase. The first digit of the Status-Code defines the class of response. The last two digits do not have any categorization role. There are 5 values for the first digit: o 1xx: Informational - Request received, continuing process o 2xx: Success - The action was successfully received, understood, and accepted o 3rr: Redirection - Further action needs to be taken in order to complete the request o 4xx: Client Error - The request contains bad syntax or cannot be fulfilled o 5xx: Server Error - The server failed to fulfill an apparently valid request The individual values of the numeric status codes defined for RTSP/2.0, and an example set of corresponding Reason-Phrases, are presented in table 4. The reason phrases listed here are only recommended; they may be replaced by local equivalents without affecting the protocol. Note that RTSP adopts most HTTP/1.1 [3] status codes and adds RTSP-specific status codes starting at x50 to avoid conflicts with newly defined HTTP status codes. RTSP status codes are extensible. RTSP applications are not required to understand the meaning of all registered status codes, though such understanding is obviously desirable. However, applications MUST understand the class of any status code, as indicated by the first digit, and treat any unrecognized response as being equivalent to the x00 status code of that class, with the exception that an unrecognized response MUST NOT be cached. For example, if an unrecognized status code of 431 is received by the client, it can safely assume that there was something wrong with its request and treat the response as if it had received a 400 status code. In such cases, user agents SHOULD present to the user the entity returned with the response, since that entity is likely to include human- H. Schulzrinne et. al. [Page 33] Internet Draft RTSP October 24, 2005 readable information which will explain the unusual status. 7.1.2 Response Header Fields The response-header fields allow the request recipient to pass additional information about the response which cannot be placed in the Status-Line. These header fields give information about the server and about further access to the resource identified by the Request-URI. All headers currently being classified as response headers are listed in table 5. Response-header field names can be extended reliably only in combination with a change in the protocol version. However, new or experimental header fields MAY be given the semantics of response- header fields if all parties in the communication recognize them to be response-header fields. Unrecognized header fields are treated as entity-header fields. 8 Entity Request and Response messages MAY transfer an entity if not otherwise restricted by the request method or response status code. An entity consists of entity-header fields and an entity-body, although some responses will only include the entity-headers. The SET_PARAMETER, and GET_PARAMETER request and response, and DESCRIBE response MAY have an entity. All 4xx and 5xx responses MAY also have an entity. In this section, both sender and recipient refer to either the client or the server, depending on who sends and who receives the entity. 8.1 Entity Header Fields Entity-header fields define optional meta-information about the entity-body or, if no body is present, about the resource identified by the request. The entity header fields are listed in table 8.1. The extension-header mechanism allows additional entity-header fields to be defined without changing the protocol, but these fields cannot be assumed to be recognizable by the recipient. Unrecognized header fields SHOULD be ignored by the recipient and forwarded by proxies. 8.2 Entity Body H. Schulzrinne et. al. [Page 34] Internet Draft RTSP October 24, 2005 See [H7.2] with the addition that an RTSP message with an entity body MUST include the Content-Type and Content-Length headers. 9 Connections RTSP requests can be transmitted over two different connection scenarios listed below: o persistent - transport connections used for several request/response transactions; o transient - transport connections used for a single request/response transaction. RFC 2326 attempted to specify an optional mechanism for transmitting RTSP messages in connectionless mode over a transport protocol such as UDP. However, it was not specified in sufficient enough detail to allow for interoperable implementations. In an attempt to reduce complexity and scope, and due to lack of interest, RTSP 2.0 does not attempt to define a mechanism for supporting RTSP over UDP or other connectionless transport protocols. A side-effect is that RTSP requests SHALL NOT be sent to multicast groups since no connection can be established with a specific receiver in multicast environments. Certain RTSP headers, such as the CSeq header (Section 14.19), which may appear to be relevant to only connectionless transport scenarios are still retained and must be implemented according to the specification. In the case of CSeq it is quite useful in proxy situations for keeping track of the different request when aggregating several client requests to a single TCP connection. 9.1 Reliability and Acknowledgements Since RTSP is transmitted primarily over connection oriented, reliable transport protocols, all RTSP requests MUST be acknowledged by the receiver. RTSP requests that are not immediately acknowledged MUST NOT be retransmitted at the application level. Instead, the application must rely on the underlying transport to provide reliability. If both the underlying reliable transport such as TCP and the RTSP application retransmit requests, each packet loss or message loss may result in two retransmissions. The receiver typically cannot take advantage of the application-layer retransmission since the transport stack will not deliver the application-layer retransmission H. Schulzrinne et. al. [Page 35] Internet Draft RTSP October 24, 2005 Code Reason Method __________________________________________________________ 100 Continue all __________________________________________________________ 200 OK all 201 Reserved n/a 250 Reserved n/a __________________________________________________________ 300 Multiple Choices all 301 Moved Permanently all 302 Found all 303 See Other all 305 Use Proxy all __________________________________________________________ 400 Bad Request all 401 Unauthorized all 402 Payment Required all 403 Forbidden all 404 Not Found all 405 Method Not Allowed all 406 Not Acceptable all 407 Proxy Authentication Required all 408 Request Timeout all 410 Gone all 411 Length Required all 412 Precondition Failed DESCRIBE, SETUP 413 Request Entity Too Large all 414 Request-URI Too Long all 415 Unsupported Media Type all 451 Parameter Not Understood SET_PARAMETER 452 reserved n/a 453 Not Enough Bandwidth SETUP 454 Session Not Found all 455 Method Not Valid In This State all 456 Header Field Not Valid all 457 Invalid Range PLAY, PAUSE 458 Parameter Is Read-Only SET_PARAMETER 459 Aggregate Operation Not Allowed all 460 Only Aggregate Operation Allowed all 461 Unsupported Transport all 462 Destination Unreachable all 463 Destination Prohibited SETUP 470 Connection Authorization Required all 471 Connection Credentials not accepted all __________________________________________________________ 500 Internal Server Error all 501 Not Implemented all 502 Bad Gateway all 503 Service Unavailable all 504 Gateway Timeout all H. Schulzrinne et. al. [Page 36] Internet Draft RTSP October 24, 2005 Table 4: Status codes and their usage with RTSP methods Header Defined in Section __________________________________________ Accept-Ranges Section 14.5 Connection-Credentials Section 14.12 ETag Section 14.21 Location Section 14.29 Proxy-Authenticate Section 14.30 Public Section 14.33 Range Section 14.34 Retry-After Section 14.36 RTP-Info Section 14.38 Scale Section 14.39 Session Section 14.42 Server Section 14.41 Speed Section 14.40 Transport Section 14.45 Unsupported Section 14.46 Vary Section 14.48 WWW-Authenticate Section 14.50 Table 5: The RTSP response headers Header Defined in Section ____________________________________ Allow Section 14.6 Content-Base Section 14.13 Content-Encoding Section 14.14 Content-Language Section 14.15 Content-Length Section 14.16 Content-Location Section 14.17 Content-Type Section 14.18 Expires Section 14.22 Last-Modified Section 14.28 Table 6: The RTSP entity headers before the first attempt has reached the receiver. If the packet loss is caused by congestion, multiple retransmissions at different layers will exacerbate the congestion. Lack of acknowledgement of an RTSP request should be handled within the constraints of the connection timeout considerations described H. Schulzrinne et. al. [Page 37] Internet Draft RTSP October 24, 2005 below (Section 9.4). 9.2 Using Connections A TCP transport can be used for both persistent connections (for several message exchanges) and transient connections (for a single message exchange). Implementations of this specification MUST support RTSP over TCP. The scheme of the RTSP URI (Section 3.2) indicates the default port that the server will listen on. A server MUST handle both persistent and transient connections. Transient connections facilitate mechanisms for fault tolerance. They also allow for application layer mobility. A server and client pair that support transient connections can survive the loss of a TCP connection, e.g. due to a NAT timeout. When the client has discovered that the TCP connection has been lost, it can set up a new one when there is need to communicate again. A persistent connection MAY be used for all transactions between the server and client, including messages to multiple RTSP sessions. However a persistent connection MAY also be closed after a few message exchanges. For example, a client may use a persistent connection for the initial SETUP and PLAY message exchanges in a session and then close the connection. Later, when the client wishes to send a new request, such as a PAUSE for the session, a new connection would be opened. This connection may either be transient or persistent. A client SHOULD NOT have more than one connection to the server at any given point. If a client or proxy handles multiple RTSP sessions on the same server, it SHOULD use only one connection for managing those sessions. This saves connection resources on the server. It also reduces complexity by and enabling the server to maintain less state about its sessions and connections. Unlike HTTP, RTSP allows a server to send requests to a client. However, this can be supported only if a client establishes a persistent connection with the server. In cases where a persistent connection does not exist between a server and its client, due to the lack of a signalling channel, the server may be forced to drop an RTSP session without notifying the client. An example of such a case is when the server desires to send a REDIRECT request for an RTSP H. Schulzrinne et. al. [Page 38] Internet Draft RTSP October 24, 2005 session to the client but is not able to do so because it cannot reach the client. Without a persistent connection between the client and the server, the media server has no reliable way of reaching the client. Also, this is the only way that requests from a server to its client are likely to traverse firewalls. In light of the above, it is RECOMMENDED that clients use persistent connections whenever possible. A client that supports persistent connections MAY "pipeline" its requests (i.e., send multiple requests without waiting for each response). A server MUST send its responses to those requests in the order that the requests were received. 9.3 Closing Connections The client MAY close a connection at any point when no outstanding request/response transactions exist for any RTSP session being managed through the connection. The server, however, SHOULD NOT close a connection until all RTSP sessions being managed through the connection have been timed out (Section 14.42). A server SHOULD NOT close a connection immediately after responding to a session-level TEARDOWN request for the last RTSP session being controlled through the connection. Instead, it should wait for a reasonable amount of time for the client to: receive the TEARDOWN response, take appropriate action, and initiate the connection closing. The server SHOULD wait at least 10 seconds after sending the TEARDOWN response before closing the connection. This is to ensure that the client has time to issue a SETUP for a new session on the existing connection after having torn the last one down. 10 seconds should give the client ample opportunity get its message to the server. A server SHOULD NOT close the connection directly as a result of responding to a request with an error code. Certain error responses such as "460 Only Aggregate Operation Allowed" (Section 13.4.12) are used for negotiating capabilities of a server with respect to content or other factors. In such cases, it is inefficient for the server to close a connection on an error response. Also, such behavior would prevent implementation of advanced/special types of requests or result in extra overhead for the client when testing for new features. On H. Schulzrinne et. al. [Page 39] Internet Draft RTSP October 24, 2005 the flip side, keeping connections open after sending an error response poses a Denial of Service security risk (Section 20). If a server initiates a connection close while the client is attempting to send a new request, the client will have to close its current connection, establish a new connection and send its request over the new connection. An RTSP message should not be terminated through a connection close. Such a message will be considered to be incomplete by the receiver and discarded. An RTSP message is properly terminated as defined in Section 4. 9.4 Timing Out Connections and RTSP Messages Receivers of a request (responder) SHOULD respond to requests in a timely manner even when a reliable transport such as TCP is used. Similarly, the sender of a request (requestor) SHOULD wait for a sufficient time for a response before concluding that the responder will not be acting upon its request. A responder SHOULD respond to all requests within 5 seconds. If the responder recognizes that processing of a request will take longer than 5 seconds, it SHOULD send a 100 response as soon as possible. It SHOULD continue sending a 100 response every 5 seconds thereafter until it is ready to send the final response to the requestor. After sending a 100 response, the receiver MUST send a final response indicating the success or failure of the request. A requestor SHOULD wait at least 10 seconds for a response before concluding that the responder will not be responding to its request. After receiving a 100 response, the requestor SHOULD continue waiting for further responses. If more than 10 seconds elapses without receiving any response, the requestor MAY assume that the responder is unresponsive and abort the connection. A requestor SHOULD wait longer than 10 seconds for a response if it is experiencing significant transport delays on its connection to the responder. The requestor is capable of determining the RTT of the request/response cycle using the Timestamp header (section 14.44) in any RTSP request. 9.5 Use of IPv6 Explicit IPv6 support was not present in RTSP 1.0 (RFC 2326). RTSP 2.0 has been updated for explicit IPv6 support. Implementations of RTSP 2.0 MUST understand literal IPv6 addresses in URIs and headers. H. Schulzrinne et. al. [Page 40] Internet Draft RTSP October 24, 2005 10 Capability Handling This section describes the capability handling mechanism available in RTSP which allows RTSP to be extended. Extensions to this version of the protocol are basically done in two ways. First, new headers can be added. Secondly, new methods can be added. The capability handling mechanism is designed to handle both cases. When a method is added, the involved parties can use the OPTIONS method to discover wether it is supported. This is done by issuing a OPTIONS request to the other party. Depending on the URI it will either apply in regards to a certain media resource, the whole server in general, or simply the next hop. The OPTIONS response will contain a Public header which declares all methods supported for the indicated resource. It is not necessary to use OPTIONS to discover support of a method, the client could simply try the method. If the receiver of the request does not support the method it will respond with an error code indicating the the method is either not implemented (501) or does not apply for the resource (405). The choice between the two discovery methods depends on the requirements of the service. Feature-Tags are defined to handle functionality additions that are not new methods. Each feature-tag represents a certain block of functionality. The amount of functionality that a feature-tag represents can vary significantly. A feature-tag can for example represent the functionality a single RTSP header provides. Another feature-tag can represent much more functionality, such as the "play.basic" feature tag which represents the minimal playback implementation. Feature-tags are used to determine wether the client, server or proxy supports the functionality that is necessary to achieve the desired service. To determine support of a feature-tag, several different headers can be used, each explained below: Supported: The supported header is used to determine the complete set of functionality that both client and server have. The intended usage is to determine before one needs to use a functionality that it is supported. It can be used in any method, however OPTIONS is the most suitable one as it at the same time determines all methods that are implemented. When sending a request the requestor declares all its capabilities by including all supported feature- tags. This results in that the receiver learns the requestors feature support. The receiver then includes its set of features in the response. H. Schulzrinne et. al. [Page 41] Internet Draft RTSP October 24, 2005 Proxy-Supported: The Proxy-Supported header is used similar to the Supported header, but instead of giving the supported functionality of the client or server it provides both the requestor and the responder a view of what functionality the proxy chain between the two supports. Proxies are required to add this header whenever the Supported header is present, but proxies may independently of the requestor add it. Require: The Require header can be included in any request where the end-point, i.e. the client or server, is required to understand the feature to correctly perform the request. This can, for example, be a SETUP request where the server is required to understand a certain parameter to be able to set up the media delivery correctly. Ignoring this parameter would not have the desired effect and is not acceptable. Therefore the end-point receiving a request containing a Require MUST negatively acknowledge any feature that it does not understand and not perform the request. The response in cases where features are not supported are 551 (Option Not Supported). Also the features that are not supported are given in the Unsupported header in the response. Proxy-Require: This method has the same purpose and workings as Require except that it only applies to proxies and not the end-point. Features that needs to be supported by both proxies and end-point needs to be included in both the Require and Proxy-Require header. Unsupported: This header is used in a 551 error response, to indicate which feature(s) that was not supported. Such a response is only the result of the usage of the Require and/or Proxy-Require header where one or more feature where not supported. This information allows the requestor to make the best of situations as it knows which features are not supported. 11 Method Definitions The method indicates what is to be performed on the resource identified by the Request-URI. The method name is case-sensitive. New methods may be defined in the future. Method names SHALL NOT start with a $ character (decimal 24) and MUST be a token as defined by the ABNF [4] in the syntax chapter 19. The methods are summarized in Table 7. H. Schulzrinne et. al. [Page 42] Internet Draft RTSP October 24, 2005 method direction object Server req. Client req. ___________________________________________________________________ DESCRIBE C -> S P,S recommended recommended GET_PARAMETER C -> S, S -> C P,S optional optional OPTIONS C -> S, S -> C P,S R=Req, Sd=Opt Sd=Req, R=Opt PAUSE C -> S P,S required required PLAY C -> S P,S required required REDIRECT S -> C P,S optional required SETUP C -> S S required required SET_PARAMETER C -> S, S -> C P,S required optional TEARDOWN C -> S P,S required required Table 7: Overview of RTSP methods, their direction, and what objects (P: presentation, S: stream) they operate on. Legend: R=Respond, Sd=Send, Opt: Optional, Req: Required, Rec: Recommended Note on Table 7: GET_PARAMETER is recommended, but not required. For example, a fully functional server can be built to deliver media without any parameters. SET_PARAMETER is required however due to its usage for keep-alive. PAUSE is now required due to that it is the only way of getting out of the state machines play state without terminating the whole session. If an RTSP agent does not support a particular method, it MUST return 501 (Not Implemented) and the requesting RTSP agent, in turn, SHOULD NOT try this method again for the given agent / resource combination. 11.1 OPTIONS The semantics of the RTSP OPTIONS method is equivalent to that of the HTTP OPTIONS method described in [H9.2]. In RTSP however, OPTIONS is bi-directional, in that a client can request it to a server and vice versa. A client MUST implement the capability to send an OPTIONS request and a server or a proxy MUST implement the capability to respond to an OPTIONS request. The client, server or proxy MAY also implement the converse of their required capability. An OPTIONS request may be issued at any time. Such a request does not modify the session state. However, it may prolong the session lifespan (see below). The URI in an OPTIONS request determines the scope of the request and the corresponding response. If the Request- URI refers to a specific media resource on a given host, the scope is limited to the set of methods supported for that media resource by the indicated RTSP agent. A Request-URI with only the host address limits the scope to the specified RTSP agent's general capabilities without regard to any specific media. If the Request-URI is an H. Schulzrinne et. al. [Page 43] Internet Draft RTSP October 24, 2005 asterisk ("*"), the scope is limited to the general capabilities of the next hop (i.e. the RTSP agent in direct communication with the request sender). Regardless of scope of the request, the Public header MUST always be included in the OPTIONS response listing the methods that are supported by the responding RTSP agent. In addition, if the scope of the request is limited to a media resource, the Allow header MUST be included in the response to enumerate the set of methods that are allowed for that resource unless the set of methods completely matches the set in the Public header. If the given resource is not available, the RTSP agent SHOULD return an appropriate response code such as 3rr or 4xx. The Supported header MAY be included in the request to query the set of features that are supported by the responding RTSP agent. The OPTIONS method can be used to keep an RTSP session alive. However, it is not the preferred means of session keep-alive signalling, see section 14.42. An OPTIONS request intended for keeping alive an RTSP session MUST include the Session header with the associated session ID. Such a request SHOULD also use the media or the aggregated control URI as the Request-URI. Example: C->S: OPTIONS * RTSP/2.0 CSeq: 1 User-Agent: PhonyClient/1.2 Require: Proxy-Require: gzipped-messages Supported: play.basic S->C: RTSP/2.0 200 OK CSeq: 1 Public: DESCRIBE, SETUP, TEARDOWN, PLAY, PAUSE Supported: play.basic, implicit-play, gzipped-messages Server: PhonyServer/1.1 Note that some of the feature-tags in Require and Proxy-Require are necessarily fictional features (one would hope that we would not purposefully overlook a truly useful feature just so that we could have a strong example in this section). 11.2 DESCRIBE H. Schulzrinne et. al. [Page 44] Internet Draft RTSP October 24, 2005 The DESCRIBE method is used to retrieve the description of a presentation or media object from a server. The Request-URI of the DESCRIBE request identifies the media resource of interest. The client MAY include the Accept header in the request to list the description formats that it understands. The server SHALL respond with a description of the requested resource and return the description in the entity of the response. The DESCRIBE reply- response pair constitutes the media initialization phase of RTSP. Example: C->S: DESCRIBE rtsp://server.example.com/fizzle/foo RTSP/2.0 CSeq: 312 User-Agent: PhonyClient 1.2 Accept: application/sdp, application/rtsl, application/mheg S->C: RTSP/2.0 200 OK CSeq: 312 Date: 23 Jan 1997 15:35:06 GMT Server: PhonyServer 1.1 Content-Type: application/sdp Content-Length: 367 v=0 o=mhandley 2890844526 2890842807 IN IP4 126.16.64.4 s=SDP Seminar i=A Seminar on the session description protocol u=http://www.example.com/lectures/sdp.ps e=seminar@example.com (Seminar Management) c=IN IP4 224.2.17.12/127 t=2873397496 2873404696 a=recvonly m=audio 3456 RTP/AVP 0 m=video 2232 RTP/AVP 31 m=application 32416 UDP WB a=orient:portrait The DESCRIBE response SHOULD contain all media initialization information for the resource(s) that it describes. Servers SHOULD NOT use the DESCRIBE response as a means of media indirection by having the description point at another server, instead usage of 3rr responses are recommended. By forcing a DESCRIBE response to contain all media H. Schulzrinne et. al. [Page 45] Internet Draft RTSP October 24, 2005 initialization for the set of streams that it describes, and discouraging the use of DESCRIBE for media indirection, any looping problems can be avoided that might have resulted from other approaches. Media initialization is a requirement for any RTSP-based system, but the RTSP specification does not dictate that this is required to be done via the DESCRIBE method. There are three ways that an RTSP client may receive initialization information: o via an RTSP DESCRIBE request o via some other protocol (HTTP, email attachment, etc.) o via some form of a user interface If a client obtains a valid description from an alternate source, the client MAY use this description for initialization purposes without issuing a DESCRIBE request for the same media. It is RECOMMENDED that minimal servers support the DESCRIBE method, and highly recommended that minimal clients support the ability to act as "helper applications" that accept a media initialization file from a user interface, and/or other means that are appropriate to the operating environment of the clients. 11.3 SETUP The SETUP request for an URI specifies the transport mechanism to be used for the streamed media. The SETUP method may be used in three different cases; Create an RTSP session, add a media to a session, and change the transport parameters of already set up media stream. When in PLAY state, using SETUP to create or add media to a session when in PLAY state is unspecified. Otherwise SETUP can be used in all three states; INIT, and READY, for both purposes and in PLAY to change the transport parameters. The Transport header, see section 14.45, specifies the transport parameters acceptable to the client for data transmission; the response will contain the transport parameters selected by the server. This allows the client to enumerate in priority order the transport mechanisms and parameters acceptable to it, while the server can select the most appropriate. It is expected that the session description format used will enable the client to select a limited number possible configurations that are offered to the server to choose from. All transport parameters SHOULD be included in the Transport header, the use of other headers for this purpose is discouraged due to middle boxes such as firewalls, or NATs. H. Schulzrinne et. al. [Page 46] Internet Draft RTSP October 24, 2005 For the benefit of any intervening firewalls, a client SHOULD indicate the transport parameters even if it has no influence over these parameters, for example, where the server advertises a fixed multicast address. Since SETUP includes all transport initialization information, firewalls and other intermediate network devices (which need this information) are spared the more arduous task of parsing the DESCRIBE response, which has been reserved for media initialization. In a SETUP response the server SHOULD include the Accept-Ranges header (see section 14.5 to indicate which time formats that are acceptable to use for this media resource. C->S: SETUP rtsp://example.com/foo/bar/baz.rm RTSP/2.0 CSeq: 302 Transport: RTP/AVP;unicast;dest_addr=":4588"/":4589", RTP/AVP/TCP;unicast;interleaved=0-1 S->C: RTSP/2.0 200 OK CSeq: 302 Date: 23 Jan 1997 15:35:06 GMT Server: PhonyServer 1.1 Session: 47112344;timeout=60 Transport: RTP/AVP;unicast;dest_addr=":4588"/":4589"; src_addr="192.0.2.241:6256"/"192.0.2.241:6257"; ssrc=2A3F93ED Accept-Ranges: NPT In the above example the client wants to create an RTSP session containing the media resource "rtsp://example.com/foo/bar/baz.rm". The transport parameters acceptable to the client is either RTP/AVP/UDP (UDP per default) to be received on client port 4588 and 4589 or RTP/AVP interleaved on the RTSP control channel. The server selects the RTP/AVP/UDP transport and adds the ports it will send and received RTP and RTCP from, and the RTP SSRC that will be used by the server. The server MUST generate a session identifier in response to a successful SETUP request, unless a SETUP request to a server includes a session identifier, in which case the server MUST bundle this setup request into the existing session (aggregated session) or return error 459 (Aggregate Operation Not Allowed) (see Section 13.4.11). H. Schulzrinne et. al. [Page 47] Internet Draft RTSP October 24, 2005 An Aggregate control URI MUST be used to control an aggregated session. This URI MUST be different from the stream control URIs of the individual media streams included in the aggregate. The Aggregate control URI is to be specified by the session description if the server supports aggregated control and aggregated control is desired for the session. However even if aggregated control is offered the client MAY chose to not set up the session in aggregated control. If an Aggregate control URI is not specified in the session description, it is normally an indication that non-aggregated control should be used. The SETUP of media streams in an aggregate which has not been given an aggregated control URI is unspecified. While the session ID sometimes has enough information for aggregate control of a session, the Aggregate control URI is still important for some methods such as SET_PARAMETER where the control URI enables the resource in question to be easily identified. The Aggregate control URI is also useful for proxies, enabling them to route the request to the appropriate server, and for logging, where it is useful to note the actual resource that a request was operating on. A session will exist until it is either removed by a TEARDOWN request or is timed-out by the server. The server MAY remove a session that has not demonstrated liveness signs from the client(s) within a certain timeout period. The default timeout value is 60 seconds; the server MAY set this to a different value and indicate so in the timeout field of the Session header in the SETUP response. For further discussion see section 14.42. Signs of liveness for an RTSP session are: o Any RTSP request from a client(s) which includes a Session header with that session's ID. o If RTP is used as a transport for the underlying media streams, an RTCP sender or receiver report from the client(s) for any of the media streams in that RTSP session. RTCP Sender Reports may for example be received in sessions where the server is invited into a conference session and is as valid for keep-alive. If a SETUP request on a session fails for any reason, the session state, as well as transport and other parameters for associated streams SHALL remain unchanged from their values as if the SETUP request had never been received by the server. 11.3.1 Changing Transport Parameters H. Schulzrinne et. al. [Page 48] Internet Draft RTSP October 24, 2005 A client MAY issue a SETUP request for a stream that is already set up or playing in the session to change transport parameters, which a server MAY allow. If it does not allow changing of parameters, it MUST respond with error 455 (Method Not Valid In This State). Reasons to support changing transport parameters, is to allow for application layer mobility and flexibility to utilize the best available transport as it becomes available. If a client receives a 455 when trying to change transport parameters while the server is in play state, it MAY try to put the server in ready state using PAUSE. Before trying issuing the SETUP request again. If also that fails the changing of transport parameters will require that the client performs a TEARDOWN of the affected media and then setting it up again. In aggregated session avoiding tearing down all the media at the same time will avoid the creation of a new session. All transport parameters MAY be changed. However the primary usage expected is to either change transport protocol completely, like switching from Interleaved TCP mode to RTP or vise versa or change delivery address. In a SETUP response for a request to change the transport parameters while in Play state, the server SHOULD include the Range to indicate from what point the new transport parameters are used. Further, if RTP is used for delivery, the server SHOULD also include the RTP-Info header to indicate from what timestamp and RTP sequence number the change has taken place. If both RTP-Info and Range is included in the response the "rtp_time" parameter and range MUST be for the corresponding time, i.e. be used in the same way as for PLAY to ensure the correct synchronization information is available. If the transport parameters change while in PLAY state results in a change of synchronization related information, for example changing RTP SSRC, the server MUST provide in the SETUP response the necessary synchronization information. However the server is RECOMMENDED to avoid changing the synchronization information if possible. 11.4 PLAY The PLAY method tells the server to start sending data via the mechanism specified in SETUP. A client MUST NOT issue a PLAY request until any outstanding SETUP requests have been acknowledged as successful. PLAY requests are valid when the session is in READY or PLAY states. A PLAY request MUST include a Session header to indicate which session the request applies to. In an aggregated session the PLAY request MUST contain an aggregated control URI. A server SHALL responde with error 460 (Only Aggregate Operation Allowed) if the client PLAY Request-URI is for one of the H. Schulzrinne et. al. [Page 49] Internet Draft RTSP October 24, 2005 media. The media in an aggregate SHALL be played in sync. If a client want individual control of the media it needs to use separate RTSP sessions for each media. The PLAY request SHALL position the normal play time to the beginning of the range specified by the Range header and delivers stream data until the end of the range if given, else to the end of the media is reached. To allow for precise composition multiple ranges MAY be specified in one PLAY Request. The range values are valid if all given ranges are part of any media within the aggregate. If a given range value points outside of the media, the response SHALL be the 457 (Invalid Range) error code. The below example will first play seconds 10 through 15, then, immediately following, seconds 20 to 25, and finally seconds 30 through the end. C->S: PLAY rtsp://audio.example.com/audio RTSP/2.0 CSeq: 835 Session: 12345678 Range: npt=10-15, npt=20-25, npt=30- See the description of the PAUSE request for further examples. A PLAY request without a Range header is legal. It SHALL start playing a stream from the beginning (npt=0-) unless the stream has been paused or is currently playing. If a stream has been paused via PAUSE, stream delivery resumes at the pause point. If a stream is currently playing, the new PLAY begins at the current stream position. The stream SHALL play until the end of the media. The Range header MUST NOT contain a time parameter. The usage of time in PLAY method has been deprecated. If a request with time parameter is received the server SHOULD respond with a 457 (Invalid Range) to indicate that the time parameter is not supported. Server MUST include a "Range" header in any PLAY response. The response MUST use the same format as the request's range header contained. If no Range header was in the request, the NPT time format SHOULD be used unless the client showed support for an other format more appropriate. Also for a session with live media streams the Range header MUST indicate a valid time. It is RECOMMENDED that normal play time is used, either the "now" indicator, for example "npt=now-", or the time since session start as an open interval, e.g. "npt=96.23-". An absolute time value (clock) for the corresponding H. Schulzrinne et. al. [Page 50] Internet Draft RTSP October 24, 2005 time MAY be given, i.e. "clock=20030213T143205Z-". The UTC clock format SHOULD only be used if client has shown support for it. A media server only supporting playback MUST support the npt format and MAY support the clock and smpte formats. For an on-demand stream, the server MUST reply with the actual range that will be played back, i.e. for which duration any media (having content at this time) is delivered. This may differ from the requested range if alignment of the requested range to valid frame boundaries is required for the media source. Note that some media streams in an aggregate may need to be delivered from even earlier points. Also, some media format have a very long duration per individual data unit, therefore it might be necessary for the client to parse the data unit, and select where to start. Example: Single audio stream (MIDI) C->S: PLAY rtsp://example.com/audio RTSP/2.0 CSeq: 836 Session: 12345678 Range: npt=7.05- S->C: RTSP/2.0 200 OK CSeq: 836 Date: 23 Jan 1997 15:35:06 GMT Server: PhonyServer 1.0 Range: npt=3.52- RTP-Info:url="rtsp://example.com/audio" ssrc=0D12F123:seq=14783;rtptime=2345962545 S->C: RTP Packet TS=2345962545 => NPT=3.52 Duration: 4.15 seconds In this example the client receives the first media packet that stretches all the way up and past the requested playtime. Thus, it is the client's decision if to render to the user the time between 3.52 and 7.05, or to skip it. In most cases it is probably most suitable to not render that time period. For live media sources it might be impossible to specify from which point in time all media streams carrying active content can actually be delivered. Therefore a server MAY specify a start time (or now-) in the range header, for which not all media will be available from. H. Schulzrinne et. al. [Page 51] Internet Draft RTSP October 24, 2005 If no range is specified in the request, the start position SHALL still be returned in the reply. If the medias that are part of an aggregate has different lengths, the PLAY request SHALL be performed as long as the given range is valid for any media, for example the longest media. Media will be sent whenever it is available for the given play-out point. A PLAY response MAY include a header(s) carrying synchronization information. As the information necessary is dependent on the media transport format, further rules specifying the header and its usage is needed. For RTP the RTP-Info header is specified, see section 14.38. After playing the desired range, the presentation does NOT transition to the READY state, media delivery simply stops. A PAUSE request MUST be issued before the stream enters the READY state. A PLAY request while the stream is still in the PLAYING state is legal, and can be issued without an intervening PAUSE request. Such a request SHALL replace the current PLAY action with the new one requested, i.e. being handle the same as the request was received in ready state. In the case the first time range in Range header has a open start time (-endtime), the server SHALL continue to play from where it currently was. A client desiring to play the media from the beginning MUST send a PLAY request with a Range header pointing at the beginning, e.g. npt=0-. If a PLAY request is received without a Range header when media delivery has stopped at the end, the server SHOULD respond with a 457 "Invalid Range" error response. In that response the current pause point in a Range header SHALL be included. The following example plays the whole presentation starting at SMPTE time code 0:10:20 until the end of the clip. Note: The RTP-Info headers has been broken into several lines to fit the page. C->S: PLAY rtsp://audio.example.com/twister.en RTSP/2.0 CSeq: 833 Session: 12345678 Range: smpte=0:10:20- S->C: RTSP/2.0 200 OK CSeq: 833 Date: 23 Jan 1997 15:35:06 GMT Server: PhonyServer 1.0 Range: smpte=0:10:22-0:15:45 RTP-Info:url="rtsp://example.com/twister.en" ssrc=0D12F123:seq=14783;rtptime=2345962545 H. Schulzrinne et. al. [Page 52] Internet Draft RTSP October 24, 2005 For playing back a recording of a live presentation, it may be desirable to use clock units: C->S: PLAY rtsp://audio.example.com/meeting.en RTSP/2.0 CSeq: 835 Session: 12345678 Range: clock=19961108T142300Z-19961108T143520Z S->C: RTSP/2.0 200 OK CSeq: 835 Date: 23 Jan 1997 15:35:06 GMT Server:PhonyServer 1.0 Range: clock=19961108T142300Z-19961108T143520Z RTP-Info:url="rtsp://example.com/meeting.en" ssrc=0D12F123:seq=53745;rtptime=484589019 All range specifiers in this specification allow for ranges with unspecified begin times (e.g. "npt=-30"). When used in a PLAY request, the server treats this as a request to start/resume playback from the current pause point, ending at the end time specified in the Range header. If the pause point is located later than the given end value, a 457 (Invalid Range) response SHALL be given. The possibility to replace a current PLAY request with a new one replaces two RTSP 1.0 functions: o The queued play functionality described in RFC 2326 [24] is removed and multiple ranges can be used to achieve a similar functionality. o The use of PLAY for keep-alive signaling, i.e. PLAY request without a range header in PLAY state, has also been deprecated. Instead a client can use, SET_PARAMETER (recommended) or OPTIONS (allowed) for keep alive. 11.5 PAUSE The PAUSE request causes the stream delivery to be interrupted (halted) temporarily. A PAUSE request MUST be done with the aggregated control URI for aggregated sessions, resulting in all media being halted, or the media URI for non-aggregated sessions. Any attempt to do muting of a single media with an PAUSE request in an aggregated session SHALL be responded with error 460 (Only Aggregate Operation Allowed). After resuming playback, H. Schulzrinne et. al. [Page 53] Internet Draft RTSP October 24, 2005 synchronization of the tracks MUST be maintained. Any server resources are kept, though servers MAY close the session and free resources after being paused for the duration specified with the timeout parameter of the Session header in the SETUP message. Example: C->S: PAUSE rtsp://example.com/fizzle/foo RTSP/2.0 CSeq: 834 Session: 12345678 S->C: RTSP/2.0 200 OK CSeq: 834 Date: 23 Jan 1997 15:35:06 GMT Range: npt=45.76- The PAUSE request MAY contain a Range header specifying when the stream or presentation is to be halted. This point is referred to as the "pause point". The time parameter in the Range MUST NOT be used. The Range header MUST contain a single value, expressed as the beginning value an open range. For example, the following clip will be played from 10 seconds through 21 seconds of the clip's normal play time, under the assumption that the PAUSE request reaches the server within 11 seconds of the PLAY request. Note that some lines has been broken in an non-correct way to fit the page: C->S: PLAY rtsp://example.com/fizzle/foo RTSP/2.0 CSeq: 834 Session: 12345678 Range: npt=10-30 S->C: RTSP/2.0 200 OK CSeq: 834 Date: 23 Jan 1997 15:35:06 GMT Server: PhonyServer 1.0 Range: npt=10-30 RTP-Info:url="rtsp://example.com/fizzle/audiotrack" ssrc=0D12F123:seq=5712;rtptime=934207921, url="rtsp://example.com/fizzle/videotrack" ssrc=4FAD8726:seq=57654;rtptime=2792482193 Session: 12345678 C->S: PAUSE rtsp://example.com/fizzle/foo RTSP/2.0 CSeq: 835 H. Schulzrinne et. al. [Page 54] Internet Draft RTSP October 24, 2005 Session: 12345678 Range: npt=21- S->C: RTSP/2.0 200 OK CSeq: 835 Date: 23 Jan 1997 15:35:09 GMT Server: PhonyServer 1.0 Range: npt=21- Session: 12345678 The pause request becomes effective the first time the server is encountering the time point specified in any of the multiple ranges. If the Range header specifies a time outside any range from the PLAY request, the error 457 (Invalid Range) SHALL be returned. If a media unit (such as an audio or video frame) starts presentation at exactly the pause point, it is not played. If the Range header is missing, stream delivery is interrupted immediately on receipt of the message and the pause point is set to the current normal play time. However, the pause point in the media stream MUST be maintained. A subsequent PLAY request without Range header SHALL resume from the pause point and play until media end. If the server has already sent data beyond the time specified in the PAUSE request's Range header, a PLAY without range SHALL resume at the point in time specified by the PAUSE request's Range header, as it is assumed that the client has discarded data after that point. This ensures continuous pause/play cycling without gaps. The pause point after any PAUSE request SHALL be returned to the client by adding a Range header with what remains unplayed of the PLAY request's ranges, i.e. including all the remaining ranges part of multiple range specification. If one desires to resume playing a ranged request, one simply includes the Range header from the PAUSE response. For example, if the server have a play request for ranges 10 to 15 and 20 to 29 pending and then receives a pause request for NPT 21, it would start playing the second range and stop at NPT 21. If the pause request is for NPT 12 and the server is playing at NPT 13 serving the first play request, the server stops immediately. If the pause request is for NPT 16, the server returns a 457 error message. To prevent that the second range is played and the server stops after completing the first range, a PAUSE request for NPT 20 needs to be issued. As another example, if a server has received requests to play ranges H. Schulzrinne et. al. [Page 55] Internet Draft RTSP October 24, 2005 10 to 15 and then 13 to 20 (that is, overlapping ranges), the PAUSE request for NPT=14 would take effect while the server plays the first range, with the second range effectively being ignored, assuming the PAUSE request arrives before the server has started playing the second, overlapping range. Regardless of when the PAUSE request arrives, it sets the pause point to 14. The below example messages is for the above case when the PAUSE request arrives before the first occurrence of NPT=14. C->S: PLAY rtsp://example.com/fizzle/foo RTSP/2.0 CSeq: 834 Session: 12345678 Range: npt=10-15, npt=13-20 S->C: RTSP/2.0 200 OK CSeq: 834 Date: 23 Jan 1997 15:35:06 GMT Server: PhonyServer 1.0 Range: npt=10-15, npt=13-20 RTP-Info:url="rtsp://example.com/fizzle/audiotrack" ssrc=0D12F123:seq=5712;rtptime=934207921, url="rtsp://example.com/fizzle/videotrack" ssrc=789DAF12:seq=57654;rtptime=2792482193 Session: 12345678 C->S: PAUSE rtsp://example.com/fizzle/foo RTSP/2.0 CSeq: 835 Session: 12345678 Range: npt=14- S->C: RTSP/2.0 200 OK CSeq: 835 Date: 23 Jan 1997 15:35:09 GMT Server: PhonyServer 1.0 Range: npt=14-15, npt=13-20 Session: 12345678 If a client issues a PAUSE request and the server acknowledges and enters the READY state, the proper server response, if the player issues another PAUSE, is still 200 OK. The 200 OK response MUST include the Range header with the current pause point, even if the PAUSE request is asking for some other pause point. See examples below: Examples: H. Schulzrinne et. al. [Page 56] Internet Draft RTSP October 24, 2005 C->S: PAUSE rtsp://example.com/fizzle/foo RTSP/2.0 CSeq: 834 Session: 12345678 S->C: RTSP/2.0 200 OK CSeq: 834 Session: 12345678 Date: 23 Jan 1997 15:35:06 GMT Range: npt=45.76-98.36 C->S: PAUSE rtsp://example.com/fizzle/foo RTSP/2.0 CSeq: 835 Session: 12345678 Range: 86- S->C: RTSP/2.0 200 OK CSeq: 835 Session: 12345678 Date: 23 Jan 1997 15:35:07 GMT Range: npt=45.76-98.36 11.6 TEARDOWN The TEARDOWN client to server request stops the stream delivery for the given URI, freeing the resources associated with it. A TEARDOWN request MAY be performed on either an aggregated or a media control URI. However some restrictions apply depending on the current state. The TEARDOWN request SHALL contain a Session header indicating what session the request applies to. A TEARDOWN using the aggregated control URI or the media URI in a session under non-aggregated control MAY be done in any state (Ready, and Play). A successful request SHALL result in that media delivery is immediately halted and the session state is destroyed. This SHALL be indicated through the lack of a Session header in the response. A TEARDOWN using a media URI in an aggregated session MAY only be done in Ready state. Such a request only removes the indicated media stream and associated resources from the session. This may result in that a session returns to non-aggregated control, due to that it only contains a single media after the requests completion. A session that will exist after the processing of the TEARDOWN request SHALL in the response to that TEARDOWN request contain a Session header. Thus the presence of the Session indicates to the receiver of the response if the session is still existing or has been removed. H. Schulzrinne et. al. [Page 57] Internet Draft RTSP October 24, 2005 Example: C->S: TEARDOWN rtsp://example.com/fizzle/foo RTSP/2.0 CSeq: 892 Session: 12345678 S->C: RTSP/2.0 200 OK CSeq: 892 Server: PhonyServer 1.0 11.7 GET_PARAMETER The GET_PARAMETER request retrieves the value of a parameter or parameters for a presentation or stream specified in the URI. If the Session header is present in a request, the value of a parameter MUST be retrieved in the specified session context. The content of the reply and response is left to the implementation. The method MAY also be used without a body (entity). If the this request is successful, i.e. a 200 OK response is received, then the keep-alive timer has been updated. Any non-required header present in such a request may or may not been processed. To allow a client to determine if any such header has been processed, it is necessary to use a feature tag and the Require header. Due to this reason it is RECOMMENDED that any parameters to be retrieved are sent in the body, rather than using any header. Example: S->C: GET_PARAMETER rtsp://example.com/fizzle/foo RTSP/2.0 CSeq: 431 Content-Type: text/parameters Session: 12345678 Content-Length: 26 packets_received jitter C->S: RTSP/2.0 200 OK CSeq: 431 Content-Length: 38 Content-Type: text/parameters packets_received: 10 jitter: 0.3838 H. Schulzrinne et. al. [Page 58] Internet Draft RTSP October 24, 2005 The "text/parameters" section is only an example type for a body carrying parameters. 11.8 SET_PARAMETER This method requests to set the value of a parameter or a set of parameters for a presentation or stream specified by the URI. The method MAY also be used without a body (entity). It is the RECOMMENDED method to use in request sent for the sole purpose of updating the keep-alive timer. If this request is successful, i.e. a 200 OK response is received, then the keep-alive timer has been updated. Any non-required header present in such a request may or may not been processed. To allow a client to determine if any such header has been processed, it is necessary to use a feature tag and the Require header. Due to this reason it is RECOMMENDED that any parameters are sent in the body, rather than using any header. A request is RECOMMENDED to only contain a single parameter to allow the client to determine why a particular request failed. If the request contains several parameters, the server MUST only act on the request if all of the parameters can be set successfully. A server MUST allow a parameter to be set repeatedly to the same value, but it MAY disallow changing parameter values. If the receiver of the request does not understand or cannot locate a parameter, error 451 (Parameter Not Understood) SHALL be used. In the case a parameter is not allowed to change, the error code is 458 (Parameter Is Read- Only). The response body SHOULD contain only the parameters that have errors. Otherwise no body SHALL be returned. Note: transport parameters for the media stream MUST only be set with the SETUP command. Restricting setting transport parameters to SETUP is for the benefit of firewalls. The parameters are split in a fine-grained fashion so that there can be more meaningful error indications. However, it may make sense to allow the setting of several parameters if an atomic setting is desirable. Imagine device control where the client does not want the camera to pan unless it can also tilt to the right angle at the same time. Example: C->S: SET_PARAMETER rtsp://example.com/fizzle/foo RTSP/2.0 CSeq: 421 H. Schulzrinne et. al. [Page 59] Internet Draft RTSP October 24, 2005 Content-length: 20 Content-type: text/parameters barparam: barstuff S->C: RTSP/2.0 451 Parameter Not Understood CSeq: 421 Content-length: 10 Content-type: text/parameters barparam The "text/parameters" section is only an example type for parameters. This method is intentionally loosely defined with the intention that the reply content and response content will be defined by the one desiring to use this mechanism. 11.9 REDIRECT The REDIRECT method is issued by a server to inform a client that it required to connect to another server location to access the resource indicated by the Request-URI. The presence of the Session header in a REDIRECT request indicates the scope of the request, and determines the specific semantics of the request. A REDIRECT request with a Session header has end-to-end (i.e. server to client) scope and applies only to the given session. Any intervening proxies SHOULD NOT disconnect the control channel while there are other remaining end-to-end sessions. The OPTIONAL Location header, if included in such a request, SHALL contain a complete absolute URI pointing to the resource to which the client SHOULD reconnect. Specifically, the Location SHALL NOT contain just the host and port. A client may receive a REDIRECT request with a Session header, if and only if, an end-to-end session has been established. A client may receive a REDIRECT request without a Session header at any time when it has communication or a connection established with a server. The scope of such a request is limited to the next-hop (i.e. the RTSP agent in direct communication with the server) and applies, as well, to the control connection between the next-hop RTSP agent and the server. A REDIRECT request without a Session header indicates that all sessions and pending requests being managed via the control connection MUST be redirected. The OPTIONAL Location header, if included in such a request, SHOULD contain an absolute URI H. Schulzrinne et. al. [Page 60] Internet Draft RTSP October 24, 2005 with only the host address and the OPTIONAL port number of the server to which the RTSP agent SHOULD reconnect. Any intervening proxies SHOULD do all of the following in the order listed: 1. respond to the REDIRECT request 2. disconnect the control channel from the requesting server 3. connect to the server at the given host address 4. pass the REDIRECT request to each applicable client (typically those clients with an active session or an unanswered request) Note: The proxy is responsible for accepting REDIRECT responses from its clients; these responses MUST NOT be passed on to either the original server or the redirected server. The lack of a Location header in any REDIRECT request is indicative of the server no longer being able to fulfill the current request and having no alternatives for the client to continue with its normal operation. It is akin to a server initiated TEARDOWN that applies both to sessions as well as the general connection associated with that client. When the Range header is not included in a REDIRECT request, the client SHOULD perform the redirection immediately and return a response to the server. The server can consider the session as terminated and can free any associated state after it receives the successful (2xx) response. The server MAY close the signalling connection upon receiving the response and the client SHOULD close the signalling connection after sending the 2xx response. The exception to this is when the client has several sessions on the server being managed by the given signalling connection. In this case, the client SHOULD close the connection when it has received and responded to REDIRECT requests for all the sessions managed by the signalling connection. If the OPTIONAL Range header is included in a REDIRECT request, it indicates when the redirection takes effec