Internet-Draft | moq-transport | March 2024 |
Curley, et al. | Expires 5 September 2024 | [Page] |
This document defines the core behavior for Media over QUIC Transport (MOQT), a media transport protocol designed to operate over QUIC and WebTransport, which have similar functionality. MOQT allows a producer of media to publish data and have it consumed via subscription by a multiplicity of endpoints. It supports intermediate content distribution networks and is designed for high scale and low latency distribution.¶
This note is to be removed before publishing as an RFC.¶
Discussion of this document takes place on the Media Over QUIC Working Group mailing list (moq@ietf.org), which is archived at https://mailarchive.ietf.org/arch/browse/moq/.¶
Source for this draft and an issue tracker can be found at https://github.com/moq-wg/moq-transport.¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
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."¶
This Internet-Draft will expire on 5 September 2024.¶
Copyright (c) 2024 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
Media Over QUIC Transport (MOQT) is a protocol that is optimized for the QUIC protocol [QUIC], either directly or via WebTransport [WebTransport], for the dissemination of media. MOQT utilizes a publish/subscribe workflow in which producers of media publish data in response to subscription requests from a multiplicity of endpoints. MOQT supports wide range of use-cases with different resiliency and latency (live, interactive) needs without compromising the scalability and cost effectiveness associated with content delivery networks.¶
MOQT is a generic protocol is designed to work in concert with multiple MoQ Streaming Formats. These MoQ Streaming Formats define how content is encoded, packaged, and mapped to MOQT objects, along with policies for discovery and subscription.¶
Section 4 covers protocol considerations on prioritization schemes and congestion response overall.¶
The development of MOQT is driven by goals in a number of areas - specifically latency, the robustness of QUIC, workflow efficiency and relay support.¶
HTTP Adaptive Streaming (HAS) has been successful at achieving scale although often at the cost of latency. Latency is necessary to correct for variable network throughput. Ideally live content is consumed at the same bitrate it is produced. End-to-end latency would be fixed and only subject to encoding and transmission delays. Unfortunately, networks have variable throughput, primarily due to congestion. Attempting to deliver content encoded at a higher bitrate than the network can support causes queuing along the path from producer to consumer. The speed at which a protocol can detect and respond to queuing determines the overall latency. TCP-based protocols are simple but are slow to detect congestion and suffer from head-of-line blocking. Protocols utilizing UDP directly can avoid queuing, but the application is then responsible for the complexity of fragmentation, congestion control, retransmissions, receiver feedback, reassembly, and more. One goal of MOQT is to achieve the best of both these worlds: leverage the features of QUIC to create a simple yet flexible low latency protocol that can rapidly detect and respond to congestion.¶
The parallel nature of QUIC streams can provide improvements in the face of loss. A goal of MOQT is to design a streaming protocol to leverage the transmission benefits afforded by parallel QUIC streams as well exercising options for flexible loss recovery. Applying [QUIC] to HAS via HTTP/3 has not yet yielded generalized improvements in throughput. One reason for this is that sending segments down a single QUIC stream still allows head-of-line blocking to occur.¶
Internet delivered media today has protocols optimized for ingest and separate protocols optimized for distribution. This protocol switch in the distribution chain necessitates intermediary origins which re-package the media content. While specialization can have its benefits, there are gains in efficiency to be had in not having to re-package content. A goal of MOQT is to develop a single protocol which can be used for transmission from contribution to distribution. A related goal is the ability to support existing encoding and packaging schemas, both for backwards compatibility and for interoperability with the established content preparation ecosystem.¶
An integral feature of a protocol being successful is its ability to deliver media at scale. Greatest scale is achieved when third-party networks, independent of both the publisher and subscriber, can be leveraged to relay the content. These relays must cache content for distribution efficiency while simultaneously routing content and deterministically responding to congestion in a multi-tenant network. A goal of MOQT is to treat relays as first-class citizens of the protocol and ensure that objects are structured such that information necessary for distribution is available to relays while the media content itself remains opaque and private.¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.¶
The party initiating a MoQ transport session.¶
The party accepting an incoming transport session.¶
A Client or Server.¶
An endpoint sending media over the network.¶
An endpoint receiving media over the network.¶
A raw QUIC connection or a WebTransport session.¶
Packet loss and queuing caused by degraded or overloaded networks.¶
A temporal sequence of objects. A group represents a join point in a track. See (Section 2.2).¶
An object is an addressable unit whose payload is a sequence of bytes. Objects form the base element in the MOQT model. See (Section 2.1).¶
An encoded bitstream. Tracks contain a sequential series of one or more groups and are the subscribable entity with MOQT.¶
This document uses the conventions detailed in ([RFC9000], Section 1.3) when describing the binary encoding.¶
As a quick reference, the following list provides a non normative summary of the parts of RFC9000 field syntax that are used in this specification.¶
Indicates that x is L bits long¶
Indicates that x holds an integer value using the variable-length encoding as described in ([RFC9000], Section 16)¶
Indicates that x can be any length including zero bits long. Values in this format always end on a byte boundary.¶
Indicates that x is optional and has a length of L¶
Indicates that x is repeated zero or more times and that each instance has a length of L¶
This document extends the RFC9000 syntax and with the additional field types:¶
Indicates that x consists of a variable length integer encoding as described in ([RFC9000], Section 16), followed by that many bytes of binary data¶
To reduce unnecessary use of bandwidth, variable length integers SHOULD be encoded using the least number of bytes possible to represent the required value.¶
MOQT has a hierarchical object model for data, comprised of objects, groups and tracks.¶
The basic data element of MOQT is an object. An object is an addressable unit whose payload is a sequence of bytes. All objects belong to a group, indicating ordering and potential dependencies. Section 2.2 An object is uniquely identified by its track namespace, track name, group ID, and object ID, and must be an identical sequence of bytes regardless of how or where it is retrieved. An Object can become unavailable, but it's contents MUST NOT change over time.¶
Objects are comprised of two parts: metadata and a payload. The metadata is never encrypted and is always visible to relays. The payload portion may be encrypted, in which case it is only visible to the producer and consumer. The application is solely responsible for the content of the object payload. This includes the underlying encoding, compression, any end-to-end encryption, or authentication. A relay MUST NOT combine, split, or otherwise modify object payloads.¶
A group is a collection of objects and is a sub-unit of a track (Section 2.3). Objects within a group SHOULD NOT depend on objects in other groups. A group behaves as a join point for subscriptions. A new subscriber might not want to receive the entire track, and may instead opt to receive only the latest group(s). The sender then selectively transmits objects based on their group membership.¶
A track is a sequence of groups (Section 2.2). It is the entity against which a consumer issues a subscription request. A subscriber can request to receive individual tracks starting at a group boundary, including any new objects pushed by the producer while the track is active.¶
In MOQT, every track has a track name and a track namespace associated with it. A track name identifies an individual track within the namespace.¶
A MOQT scope is a set of servers (as identified by their connection URIs) for which the tuple of Track Name and Track Namespace are guaranteed to be unique and identify a specific track. It is up to the application using MOQT to define how broad or narrow the scope is. An application that deals with connections between devices on a local network may limit the scope to a single connection; by contrast, an application that uses multiple CDNs to serve media may require the scope to include all of those CDNs.¶
Because the tuple of Track Namespace and Track Name are unique within an MOQT scope, they can be used as a cache key. MOQT does not provide any in-band content negotiation methods similar to the ones defined by HTTP ([RFC9110], Section 10); if, at a given moment in time, two tracks within the same scope contain different data, they have to have different names and/or namespaces.¶
In this specification, both the Track Namespace and the Track Name are not constrained to a specific encoding. They carry a sequence of bytes and comparison between two Track Namespaces or Track Names is done by exact comparison of the bytes. Specifications that use MoQ Transport may constrain the information in these fields, for example by restricting them to UTF-8. Any specification that does needs to specify the canonicalization into the bytes in the Track Namespace or Track Name such that exact comparison works.¶
Each track MAY have one or more associated connection URLs specifying network hosts through which a track may be accessed. The syntax of the Connection URL and the associated connection setup procedures are specific to the underlying transport protocol usage Section 3.¶
This document defines a protocol that can be used interchangeably both over a QUIC connection directly [QUIC], and over WebTransport [WebTransport]. Both provide streams and datagrams with similar semantics (see [I-D.ietf-webtrans-overview], Section 4); thus, the main difference lies in how the servers are identified and how the connection is established. When using QUIC, datagrams MUST be supported via the [QUIC-DATAGRAM] extension, which is already a requirement for WebTransport over HTTP/3.¶
There is no definition of the protocol over other transports, such as TCP, and applications using MoQ might need to fallback to another protocol when QUIC or WebTransport aren't available.¶
A MOQT server that is accessible via WebTransport can be identified using an HTTPS URI ([RFC9110], Section 4.2.2). A MOQT session can be established by sending an extended CONNECT request to the host and the path indicated by the URI, as described in [WebTransport], Section 3.¶
A MOQT server that is accessible via native QUIC can be identified by a URI with a "moq" scheme. The "moq" URI scheme is defined as follows, using definitions from [RFC3986]:¶
moq-URI = "moqt" "://" authority path-abempty [ "?" query ]¶
The authority
portion MUST NOT contain a non-empty host
portion.
The moq
URI scheme supports the /.well-known/
path prefix defined in
[RFC8615].¶
This protocol does not specify any semantics on the path-abempty
and
query
portions of the URI. The contents of those are left up to the
application.¶
The client can establish a connection to a MoQ server identified by a
given URI by setting up a QUIC connection to the host and port
identified by the authority
section of the URI. The path-abempty
and query
portions of the URI are communicated to the server using the
PATH parameter (Section 6.2.2.2) which is sent in the CLIENT_SETUP message at the
start of the session. The ALPN value [RFC7301] used by the protocol
is moq-00
.¶
Endpoints use the exchange of Setup messages to negotiate the MOQT version and any extensions to use.¶
The client indicates the MOQT versions it supports in the CLIENT_SETUP message (see Section 6.2). It also includes the union of all Setup Parameters Section 6.2.2 required for a handshake by any of those versions.¶
Within any MOQT version, clients request the use of extensions by adding Setup parameters corresponding to that extension. No extensions are defined in this document.¶
The server replies with a SERVER_SETUP message that indicates the chosen version, includes all parameters required for a handshake in that version, and parameters for every extension requested by the client that it supports.¶
New versions of MOQT MUST specify which existing extensions can be used with that version. New extensions MUST specify the existing versions with which they can be used.¶
If a given parameter carries the same information in multiple versions, but might have different optimal values in those versions, there SHOULD be separate Setup parameters for that information in each version.¶
The first stream opened is a client-initiated bidirectional control stream where the peers exchange Setup messages (Section 6.2). All messages defined in this draft except OBJECT and OBJECT_WITH_LENGTH are sent on the control stream after the Setup message. Control messages MUST NOT be sent on any other stream, and a peer receiving a control message on a different stream closes the session as a 'Protocol Violation'. Objects MUST NOT be sent on the control stream, and a peer receiving an Object on the control stream closes the session as a 'Protocol Violation'.¶
This draft only specifies a single use of bidirectional streams. Objects are sent on unidirectional streams. Because there are no other uses of bidirectional streams, a peer MAY currently close the session as a 'Protocol Violation' if it receives a second bidirectional stream.¶
The control stream MUST NOT be abruptly closed at the underlying transport layer. Doing so results in the session being closed as a 'Protocol Violation'.¶
Streams aside from the control stream MAY be canceled due to congestion or other reasons by either the sender or receiver. Early termination of a stream does not affect the MoQ application state, and therefore has no effect on outstanding subscriptions.¶
The transport session can be terminated at any point. When native QUIC is used, the session is closed using the CONNECTION_CLOSE frame ([QUIC], Section 19.19). When WebTransport is used, the session is closed using the CLOSE_WEBTRANSPORT_SESSION capsule ([WebTransport], Section 5).¶
The application MAY use any error message and SHOULD use a relevant code, as defined below:¶
Code | Reason |
---|---|
0x0 | No Error |
0x1 | Internal Error |
0x2 | Unauthorized |
0x3 | Protocol Violation |
0x4 | Duplicate Track Alias |
0x5 | Parameter Length Mismatch |
0x10 | GOAWAY Timeout |
No Error: The session is being terminated without an error.¶
Internal Error: An implementation specific error occurred.¶
Unauthorized: The endpoint breached an agreement, which MAY have been pre-negotiated by the application.¶
Protocol Violation: The remote endpoint performed an action that was disallowed by the specification.¶
Duplicate Track Alias: The endpoint attempted to use a Track Alias that was already in use.¶
GOAWAY Timeout: The session was closed because the client took too long to close the session in response to a GOAWAY (Section 6.14) message. See session migration (Section 3.6).¶
MoqTransport requires a long-lived and stateful session. However, a service provider needs the ability to shutdown/restart a server without waiting for all sessions to drain naturally, as that can take days for long-form media. MoqTransport avoids this via the GOAWAY message (Section 6.14).¶
The server sends a GOAWAY message, signaling that the client should establish a new session and migrate any active subscriptions. The GOAWAY message may contain a new URI for the new session, otherwise the current URI is reused. The server SHOULD terminate the session with 'GOAWAY Timeout' after a sufficient timeout if there are still open subscriptions on a connection.¶
The GOAWAY message does not immediately impact subscription state. A subscriber SHOULD individually UNSUBSCRIBE for each existing subscription, while a publisher MAY reject new SUBSCRIBEs while in the draining state. When the server is a subscriber, it SHOULD send a GOAWAY message prior to any UNSUBSCRIBE messages.¶
After the client receives a GOAWAY, it's RECOMMENDED that the client waits until there are no more active subscriptions before closing the session with NO_ERROR. Ideally this is transparent to the application using MOQT, which involves establishing a new session in the background and migrating active subscriptions and announcements. The client can choose to delay closing the session if it expects more OBJECTs to be delivered. The server closes the session with a 'GOAWAY Timeout' if the client doesn't close the session quickly enough.¶
TODO: This is a placeholder section to capture details on how the MOQT protocol deals with prioritization and congestion overall.¶
This section is expected to cover details on:¶
Prioritization Schemes.¶
Congestion Algorithms and impacts.¶
Mapping considerations for one object per stream vs multiple objects per stream.¶
Considerations for merging multiple streams across domains onto single connection and interactions with specific prioritization schemes.¶
At the point of this writing, the working group has not reached consensus on several important goals, such as:¶
Ensuring that objects are delivered in the order intended by the emitter¶
Allowing nodes and relays to skip or delay some objects to deal with congestion¶
Ensuring that emitters can accurately predict the behavior of relays¶
Ensuring that when relays have to skip and delay objects belonging to different tracks that they do it in a predictable way if tracks are explicitly coordinated and in a fair way if they are not.¶
The working group has been considering two alternatives: marking objects belonging to a track with an explicit "send order"; and, defining algorithms combining tracks, priorities and object order within a group. The two proposals are listed in Section 4.1.1 and Section 4.1.2. We expect further work before a consensus is reached.¶
Media is produced with an intended order, both in terms of when media should be presented (PTS) and when media should be decoded (DTS). As stated in the introduction, the network is unable to maintain this ordering during congestion without increasing latency.¶
The encoder determines how to behave during congestion by assigning each object a numeric send order. The send order SHOULD be followed when possible, to ensure that the most important media is delivered when throughput is limited. Note that the contents within each object are still delivered in order; this send order only applies to the ordering between objects.¶
A sender MUST send each object over a dedicated stream. The library should support prioritization (Section 4) such that streams are transmitted in send order.¶
A receiver MUST NOT assume that objects will be received in send order, for the following reasons:¶
Newly encoded objects can have a smaller send order than outstanding objects.¶
Packet loss or flow control can delay the send of individual streams.¶
The sender might not support stream prioritization.¶
TODO: Refer to Congestion Response and Prioritization Section for further details on various proposals.¶
Media is produced as a set of layers, such as for example low definition and high definition, or low frame rate and high frame rate. Each object belonging to a track and a group has two attributes: the object-id, and the priority (or layer).¶
When nodes or relays have to choose which object to send next, they apply the following rules:¶
within the same group, objects with a lower priority number (e.g. P1) are always sent before objects with a numerically greater priority number (e.g., P2)¶
within the same group, and the same priority level, objects with a lower object-id are always sent before objects with a higher object-id.¶
objects from later groups are normally always sent before objects of previous groups.¶
The latter rule is generally agreed as a way to ensure freshness, and to recover quickly if queues and delays accumulate during a congestion period. However, there may be cases when finishing the transmission of an ongoing group results in better user experience than strict adherence to the freshness rule. We expect that that the working group will eventually reach consensus and define meta data that controls this behavior.¶
There have been proposals to allow emitters to coordinate the allocation of layer priorities across multiple coordinated tracks. At this point, these proposals have not reached consensus.¶
Relays are leveraged to enable distribution scale in the MoQ architecture. Relays can be used to form an overlay delivery network, similar in functionality to Content Delivery Networks (CDNs). Additionally, relays serve as policy enforcement points by validating subscribe and publish requests at the edge of a network.¶
Subscribers interact with the Relays by sending a SUBSCRIBE (Section 6.4) control message for the tracks of interest. Relays MUST ensure subscribers are authorized to access the content associated with the track. The authorization information can be part of subscription request itself or part of the encompassing session. The specifics of how a relay authorizes a user are outside the scope of this specification.¶
The subscriber making the subscribe request is notified of the result of the subscription, via SUBSCRIBE_OK (Section 6.5) or the SUBSCRIBE_ERROR Section 6.6 control message. The entity receiving the SUBSCRIBE MUST send only a single response to a given SUBSCRIBE of either SUBSCRIBE_OK or SUBSCRIBE_ERROR.¶
For successful subscriptions, the publisher maintains a list of subscribers for each track. Each new OBJECT belonging to the track within the subscription range is forwarded to each active subscriber, dependent on the congestion response. A subscription remains active until the publisher of the track terminates the track with a SUBSCRIBE_DONE (see Section 6.8).¶
Objects MUST NOT be sent for unsuccessful subscriptions, and if a subscriber receives a SUBSCRIBE_ERROR after receiving objects, it MUST close the session with a 'Protocol Violation'.¶
A relay MUST not reorder or drop objects received on a multi-object stream when forwarding to subscribers, unless it has application specific information.¶
Relays MAY aggregate authorized subscriptions for a given track when multiple subscribers request the same track. Subscription aggregation allows relays to make only a single forward subscription for the track. The published content received from the forward subscription request is cached and shared among the pending subscribers.¶
The application SHOULD use a relevant error code in SUBSCRIBE_ERROR, as defined below:¶
Code | Reason |
---|---|
0x0 | Internal Error |
0x1 | Invalid Range |
0x2 | Retry Track Alias |
The applicaiton SHOULD use a relevant status code in SUBSCRIBE_DONE, as defined below:¶
Code | Reason |
---|---|
0x0 | Unsubscribed |
0x1 | Internal Error |
0x2 | Unauthorized |
0x3 | Track Ended |
0x4 | Subscription Ended |
0x5 | Going Away |
0x6 | Expired |
Publishing through the relay starts with publisher sending ANNOUNCE
control message with a Track Namespace
(Section 2.3).¶
Relays MUST ensure that publishers are authorized by:¶
Verifying that the publisher is authorized to publish the content associated with the set of tracks whose Track Namespace matches the announced namespace. Specifics of where the authorization happens, either at the relays or forwarded for further processing, depends on the way the relay is managed and is application specific (typically based on prior business agreement).¶
Relays respond with an ANNOUNCE_OK or ANNOUNCE_ERROR control message providing the result of announcement. The entity receiving the ANNOUNCE MUST send only a single response to a given ANNOUNCE of either ANNOUNCE_OK or ANNOUNCE_ERROR. When a publisher wants to stop new subscriptions for an announced namespace it sends an UNANNOUNCE. A subscriber indicates it will no longer route subscriptions for a namespace it previously responded ANNOUNCE_OK to by sending an ANNOUNCE_CANCEL.¶
A relay manages sessions from multiple publishers and subscribers, connecting them based on the track namespace. This MUST use an exact match on track namespace unless otherwise negotiated by the application. For example, a SUBSCRIBE namespace=foobar message will be forwarded to the session that sent ANNOUNCE namespace=foobar.¶
OBJECT message headers carry a short hop-by-hop Track Alias
that maps to
the Full Track Name (see Section 6.5). Relays use the
Track Alias
of an incoming OBJECT message to identify its track and find
the active subscribers for that track. Relays MUST NOT depend on OBJECT
payload content for making forwarding decisions and MUST only depend on the
fields, such as priority order and other metadata properties in the
OBJECT message header. Unless determined by congestion response, Relays
MUST forward the OBJECT message to the matching subscribers.¶
TODO: Refer to Section 4. Add details to describe relay behavior when merging or splitting streams and interactions with congestion response.¶
MOQT encodes the delivery information for a stream via OBJECT headers (Section 6.3). A relay MUST NOT modify Object properties when forwarding.¶
A relay MUST treat the object payload as opaque. A relay MUST NOT combine, split, or otherwise modify object payloads. A relay SHOULD prioritize streams (Section 4) based on the send order/priority.¶
A sender SHOULD begin sending incomplete objects when available to avoid incurring additional latency.¶
A relay that reads from a stream and writes to stream in order will introduce head-of-line blocking. Packet loss will cause stream data to be buffered in the library, awaiting in order delivery, which will increase latency over additional hops. To mitigate this, a relay SHOULD read and write stream data out of order subject to flow control limits. See section 2.2 in [QUIC].¶
Both unidirectional and bidirectional streams contain sequences of length-delimited messages.¶
An endpoint that receives an unknown message type MUST close the session.¶
ID | Messages |
---|---|
0x0 | OBJECT_STREAM (Section 6.3.2) |
0x1 | OBJECT_DATAGRAM (Section 6.3.2) |
0x3 | SUBSCRIBE (Section 6.4) |
0x4 | SUBSCRIBE_OK (Section 6.5) |
0x5 | SUBSCRIBE_ERROR (Section 6.6) |
0x6 | ANNOUNCE (Section 6.9) |
0x7 | ANNOUNCE_OK (Section 6.10) |
0x8 | ANNOUNCE_ERROR (Section 6.11) |
0x9 | UNANNOUNCE (Section 6.12) |
0xA | UNSUBSCRIBE (Section 6.7) |
0xB | SUBSCRIBE_DONE (Section 6.8) |
0x10 | GOAWAY (Section 6.14) |
0x40 | CLIENT_SETUP (Section 6.2) |
0x41 | SERVER_SETUP (Section 6.2) |
0x50 | STREAM_HEADER_TRACK (Section 6.3.3) |
0x51 | STREAM_HEADER_GROUP (Section 6.3.3) |
Some messages include a Parameters field that encode optional message elements. They contain a type, length, and value.¶
Senders MUST NOT repeat the same parameter type in a message. Receivers SHOULD check that there are no duplicate parameters and close the session as a 'Protocol Violation' if found.¶
Receivers ignore unrecognized parameters.¶
The format of Parameters is as follows:¶
Parameter Type is an integer that indicates the semantic meaning of the parameter. Setup message parameters use a namespace that is constant across all MoQ Transport versions. All other messages use a version-specific namespace. For example, the integer '1' can refer to different parameters for Setup messages and for all other message types.¶
SETUP message parameter types are defined in Section 6.2.2. Version- specific parameter types are defined in Section 6.1.1.¶
The Parameter Length field of the String Parameter encodes the length of the Parameter Value field in bytes.¶
Each parameter description will indicate the data type in the Parameter Value field. If a receiver understands a parameter type, and the parameter length implied by that type does not match the Parameter Length field, the receiver MUST terminate the session with error code 'Parameter Length Mismatch'.¶
Each version-specific parameter definition indicates the message types in which it can appear. If it appears in some other type of message, it MUST be ignored. Note that since Setup parameters use a separate namespace, it is impossible for these parameters to appear in Setup messages.¶
The CLIENT_SETUP
and SERVER_SETUP
messages are the first messages exchanged
by the client and the server; they allows the peers to establish the mutually
supported version and agree on the initial configuration before any objects are
exchanged. It is a sequence of key-value pairs called Setup parameters; the
semantics and format of which can vary based on whether the client or server is
sending. To ensure future extensibility of MOQT, the peers MUST ignore unknown
setup parameters. TODO: describe GREASE for those.¶
The wire format of the Setup messages are as follows:¶
The available versions and Setup parameters are detailed in the next sections.¶
MoQ Transport versions are a 32-bit unsigned integer, encoded as a varint. This version of the specification is identified by the number 0x00000001. Versions with the most significant 16 bits of the version number cleared are reserved for use in future IETF consensus documents.¶
The client offers the list of the protocol versions it supports; the server MUST reply with one of the versions offered by the client. If the server does not support any of the versions offered by the client, or the client receives a server version that it did not offer, the corresponding peer MUST close the session.¶
[[RFC editor: please remove the remainder of this section before publication.]]¶
The version number for the final version of this specification (0x00000001), is reserved for the version of the protocol that is published as an RFC. Version numbers used to identify IETF drafts are created by adding the draft number to 0xff000000. For example, draft-ietf-moq-transport-13 would be identified as 0xff00000D.¶
The ROLE parameter (key 0x00) allows each endpoint to independently specify what funnctionality they support for the session. It has three possible values, which are of type varint:¶
The endpoint can process subscriptions and send objects, but not subscribe. The endpoint MUST NOT send a SUBSCRIBE message and an ANNOUNCE MUST NOT be sent to it.¶
The endpoint can send subscriptions and receive objects, but not publish. The endpoint MUST NOT send an ANNOUNCE message and a SUBSCRIBE MUST NOT be sent to it.¶
The endpoint can act as a publisher or subscriber, and can send or process any message type.¶
Both endpoints MUST send a ROLE parameter with one of the three values specified above. Both endpoints MUST close the session if the ROLE parameter is missing or is not one of the three above-specified values.¶
The PATH parameter (key 0x01) allows the client to specify the path of the MoQ URI when using native QUIC ([QUIC]). It MUST NOT be used by the server, or when WebTransport is used. If the peer receives a PATH parameter from the server, or when WebTransport is used, it MUST close the connection. It follows the URI formatting rules [RFC3986].¶
When connecting to a server using a URI with the "moq" scheme, the
client MUST set the PATH parameter to the path-abempty
portion of the
URI; if query
is present, the client MUST concatenate ?
, followed by
the query
portion of the URI to the parameter.¶
An OBJECT message contains a range of contiguous bytes from from the specified track, as well as associated metadata required to deliver, cache, and forward it.¶
A canonical MoQ Object has the following information:¶
Track Namespace and Track Name: The track this object belongs to.¶
Group ID: The object is a member of the indicated group ID Section 2.2 within the track.¶
Object ID: The order of the object within the group. The IDs starts at 0, increasing sequentially for each object within the group.¶
Object Send Order: An integer indicating the object send order Section 4.1.1 or priority Section 4.1.2 value.¶
Object Forwarding Preference: An enumeration indicating how a sender sends an
object. The preferences are Track, Group, Object and Datagram. An Object MUST
be sent according to its Object Forwarding Preference
, described below.¶
Object Payload: An opaque payload intended for the consumer and SHOULD NOT be processed by a relay.¶
Every Track has a single 'Object Forwarding Preference' and publishers MUST NOT mix different forwarding preferences within a single track. If a subscriber receives different forwarding preferences for a track, it SHOULD close the session with an error of 'Protocol Violation'.¶
Object Stream Message¶
An OBJECT_STREAM
message carries a single object on a stream. There is no
explicit length of the payload; it is determined by the end of the stream. An
OBJECT_STREAM
message MUST be the first and only message on a unidirectional
stream.¶
An Object received in an OBJECT_STREAM
message has an Object Forwarding
Preference
= Object
.¶
To send an Object with Object Forwarding Preference
= Object
, open a stream,
serialize object fields below, and terminate the stream.¶
Subscribe ID: Subscription Identifer as defined in Section 6.4.¶
Track Alias: Identifies the Track Namespace and Track Name as defined in Section 6.4.¶
If the Track Namespace and Track Name identified by the Track Alias are different from those specified in the subscription identified by Subscribe ID, the receiver MUST close the session with a Protocol Violation.¶
Other fields: As described in Section 6.3.1.¶
Object Datagram Message¶
An OBJECT_DATAGRAM
message carries a single object in a datagram.
There is no explicit length of the payload; it is determined by the
length of the datagram.¶
An Object received in an OBJECT_DATAGRAM
message has an Object
Forwarding Preference
= Datagram
. To send an Object with Object
Forwarding Preference
= Datagram
, determine the length of the fields and
payload and send the Object as datagram. In certain scenarios where the object
size can be larger than maximum datagram size for the session, the Object
will be dropped.¶
When multiple objects are sent on a stream, the stream begins with a stream header message and is followed by one or more sets of serialized object fields. If a stream ends gracefully in the middle of a serialized Object, terminate the session with a Protocol Violation.¶
A sender SHOULD NOT open more than one multi-object stream at a time with the same stream header message type and fields.¶
TODO: figure out how a relay closes these streams¶
Stream Header Track¶
When a stream begins with STREAM_HEADER_TRACK
, all objects on the stream
belong to the track requested in the Subscribe message identified by Subscribe
ID
. All objects on the stream have the Object Send Order
specified in the
stream header.¶
All Objects received on a stream opened with STREAM_HEADER_TRACK have an Object
Forwarding Preference
= Track
.¶
To send an Object with Object Forwarding Preference
= Track
, find the open
stream that is associated with the subscription, or open a new one and send the
STREAM_HEADER_TRACK
if needed, then serialize the the following object fields.¶
Stream Header Group¶
A sender MUST NOT send an Object on a stream if its Group ID is less than a previously sent Group ID on that stream, or if its Object ID is less than or equal to a previously sent Object ID within a given group on that stream.¶
When a stream begins with STREAM_HEADER_GROUP
, all objects on the stream
belong to the track requested in the Subscribe message identified by Subscribe
ID
and the group indicated by Group ID
. All objects on the stream
have the Object Send Order
specified in the stream header.¶
All Objects received on a stream opened with STREAM_HEADER_GROUP
have an
Object Forwarding Preference
= Group
.¶
To send an Object with Object Forwarding Preference
= Group
, find the open
stream that is associated with the subscription, Group ID
and Object
Send Order
, or open a new one and send the STREAM_HEADER_GROUP
if needed,
then serialize the following fields.¶
A sender MUST NOT send an Object on a stream if its Object ID is less than a previously sent Object ID within a given group in that stream.¶
Sending a track on one stream:¶
STREAM_HEADER_TRACK { Subscribe ID = 1 Track Alias = 1 Object Send Order = 0 } { Group ID = 0 Object ID = 0 Object Payload Length = 4 Payload = "abcd" } { Group ID = 1 Object ID = 0 Object Payload Length = 4 Payload = "efgh" }¶
Sending a group on one stream, with a unordered object in the group appearing on its own stream.¶
Stream = 2 STREAM_HEADER_GROUP { Subscribe ID = 2 Track Alias = 2 Group ID = 0 Object Send Order = 0 } { Object ID = 0 Object Payload Length = 4 Payload = "abcd" } { Object ID = 1 Object Payload Length = 4 Payload = "efgh" } Stream = 6 OBJECT_STREAM { Subscribe ID = 2 Track Alias = 2 Group ID = 0 Object ID = 1 Payload = "moqrocks" }¶
A receiver issues a SUBSCRIBE to a publisher to request a track.¶
The receiver specifies a start and optional end Location
for the subscription.
A location value may be an absolute group or object sequence, or it may be a
delta relative to the largest group or the largest object in a group.¶
Location { Mode (i), [Value (i)], }¶
There are 4 modes:¶
None (0x0): The Location is unspecified, Value is not present¶
Absolute (0x1): Value is an absolute sequence¶
RelativePrevious (0x2): Value is a delta from the largest sequence. 0 is the largest sequence, 1 is the largest sequence - 1, and so on.¶
RelativeNext (0x3): Value is a delta from the largest sequence. 0 is the largest sequence + 1, 1 is the largest sequence + 2, and so on.¶
The following table shows an example of how the RelativePrevious and RelativeNext values are used to determine the absolute sequence.¶
The format of SUBSCRIBE is as follows:¶
Subscribe ID: The subscription identifier that is unique within the session.
Subscribe ID
is a monotonically increasing variable length integer which
MUST not be reused within a session. Subscribe ID
is used by subscribers and
the publishers to identify a given subscription. Subscribers specify the
Subscribe ID
and it is included in the corresponding SUBSCRIBE_OK or
SUBSCRIBE_ERROR messages.¶
Track Alias: A session specific identifier for the track. Messages that reference a track, such as OBJECT (Section 6.3), reference this Track Alias instead of the Track Name and Track Namespace to reduce overhead. If the Track Alias is already in use, the publisher MUST close the session with a Duplicate Track Alias error (Section 3.5).¶
Track Namespace: Identifies the namespace of the track as defined in (Section 2.3.1).¶
Track Name: Identifies the track name as defined in (Section 2.3.1).¶
StartGroup: The Location of the requested group. StartGroup's Mode MUST NOT be None.¶
StartObject: The Location of the requested object. StartObject's Mode MUST NOT be None.¶
EndGroup: The last Group requested in the subscription, inclusive. EndGroup's Mode is None for an open-ended subscription.¶
EndObject: The last Object requested in the subscription, exclusive. EndObject's Mode MUST be None if EndGroup's Mode is None. EndObject's Mode MUST NOT be None if EndGroup's Mode is not None.¶
Track Request Parameters: The parameters are defined in Section 6.1.1¶
On successful subscription, the publisher SHOULD start delivering objects from the group ID and object ID described above.¶
If a publisher cannot satisfy the requested start or end for the subscription it MAY send a SUBSCRIBE_ERROR with code 'Invalid Range'. A publisher MUST NOT send objects from outside the requested start and end.¶
TODO: Define the flow where subscribe request matches an existing subscribe id (subscription updates.)¶
1. Now Start Group: Mode=RelativePrevious, Value=0 Start Object: Mode=RelateiveNext, Value=0 End Group: Mode=None End Object: Mode=None StartGroup=Largest Group StartObject=Largest Object + 1 2. Current Start Group: Mode=RelativePrevious, Value=0 Start Object: Mode=Absolute, Value=0 End Group: Mode=None End Object: Mode=None StartGroup=Largest Group StartObject=0 3. Previous Start Group: Mode=RelativePrevious, Value=1 Start Object: Mode=Absolute, Value=0 End Group: Mode=None End Object: Mode=None StartGroup=Largest Group - 1 StartObject=0 4. Next Start Group: Mode=RelativeNext, Value=0 Start Object: Mode=Absolute, Value=0 End Group: Mode=None End Object: Mode=None StartGroup=Largest Group + 1 StartObject=0 5. Range, All of group 3 Start Group: Mode=Absolute, Value=3 Start Object: Mode=Absolute, Value=0 End Group: Mode=Absolute, Value=4 End Object: Mode=Absolute, Value=0 Start = Group 3, Object 0 End = Group 3, Object <last>¶
A SUBSCRIBE_OK control message is sent for successful subscriptions.¶
Subscribe ID: Subscription Identifer as defined in Section 6.4.¶
Expires: Time in milliseconds after which the subscription is no longer valid. A value of 0 indicates that the subscription does not expire or expires at an unknown time. Expires is advisory and a subscription can end prior to the expiry time or last longer.¶
ContentExists: 1 if an object has been published on this track, 0 if not. If 0, then the Largest Group ID and Largest Object ID fields will not be present.¶
Largest Group ID: the largest Group ID available for this track. This field is only present if ContentExists has a value of 1.¶
Largest Object ID: the largest Object ID available within the largest Group ID for this track. This field is only present if ContentExists has a value of 1.¶
A publisher sends a SUBSCRIBE_ERROR control message in response to a failed SUBSCRIBE.¶
Subscribe ID: Subscription Identifer as defined in Section 6.4.¶
Error Code: Identifies an integer error code for subscription failure.¶
Reason Phrase: Provides the reason for subscription error.¶
Track Alias: When Error Code is 'Retry Track Alias', the subscriber SHOULD re-issue the SUBSCRIBE with this Track Alias instead. If this Track Alias is already in use, the receiver MUST close the connection with a Duplicate Track Alias error (Section 3.5).¶
A subscriber issues a UNSUBSCRIBE
message to a publisher indicating it is no
longer interested in receiving media for the specified track and Objects
should stop being sent as soon as possible. The publisher sends a
SUBSCRIBE_DONE to acknowledge the unsubscribe was successful and indicate
the final Object.¶
The format of UNSUBSCRIBE
is as follows:¶
Subscribe ID: Subscription Identifer as defined in Section 6.4.¶
A publisher issues a SUBSCRIBE_DONE
message to indicate it
is done publishing Objects for that subscription. The Status Code indicates why
the subscription ended, and whether it was an error.¶
The format of SUBSCRIBE_DONE
is as follows:¶
Subscribe ID: Subscription identifier as defined in Section 6.4.¶
Status Code: An integer status code indicating why the subscription ended.¶
Reason Phrase: Provides the reason for subscription error.¶
ContentExists: 1 if an object has been published for this subscription, 0 if not. If 0, then the Final Group and Final Object fields will not be present.¶
Final Group: The largest Group ID sent by the publisher in an OBJECT message in this track.¶
Final Object: The largest Object ID sent by the publisher in an OBJECT
message in the Final Group
for this track.¶
The publisher sends the ANNOUNCE control message to advertise where the receiver can route SUBSCRIBEs for tracks within the announced Track Namespace. The receiver verifies the publisher is authorized to publish tracks under this namespace.¶
Track Namespace: Identifies a track's namespace as defined in (Section 2.3.1)¶
Parameters: The parameters are defined in Section 6.1.1.¶
The receiver sends an ANNOUNCE_OK control message to acknowledge the successful authorization and acceptance of an ANNOUNCE message.¶
Track Namespace: Identifies the track namespace in the ANNOUNCE message for which this response is provided.¶
The receiver sends an ANNOUNCE_ERROR control message for tracks that failed authorization.¶
The publisher sends the UNANNOUNCE
control message to indicate
its intent to stop serving new subscriptions for tracks
within the provided Track Namespace.¶
Track Namespace: Identifies a track's namespace as defined in (Section 2.3.1).¶
The subscriber sends an ANNOUNCE_CANCEL
control message to
indicate it will stop sending new subscriptions for tracks
within the provided Track Namespace.¶
If a publisher recieves new subscriptions for that namespace after receiving an ANNOUNCE_CANCEL, it SHOULD close the session as a 'Protocol Violation'.¶
Track Namespace: Identifies a track's namespace as defined in (Section 2.3.1).¶
The server sends a GOAWAY
message to initiate session migration
(Section 3.6) with an optional URI.¶
The server MUST terminate the session with a Protocol Violation (Section 3.5) if it receives a GOAWAY message. The client MUST terminate the session with a Protocol Violation (Section 3.5) if it receives multiple GOAWAY messages.¶
New Session URI: The client MUST use this URI for the new session if provded. If the URI is zero bytes long, the current URI is reused instead. The new session URI SHOULD use the same scheme as the current URL to ensure compatibility.¶
TODO: Expand this section, including subscriptions.¶
Live content requires significant bandwidth and resources. Failure to set limits will quickly cause resource exhaustion.¶
MOQT uses stream limits and flow control to impose resource limits at the network layer. Endpoints SHOULD set flow control limits based on the anticipated bitrate.¶
Endpoints MAY impose a MAX STREAM count limit which would restrict the number of concurrent streams which a MOQT Streaming Format could have in flight.¶
The producer prioritizes and transmits streams out of order. Streams might be starved indefinitely during congestion. The producer and consumer MUST cancel a stream, preferably the lowest priority, after reaching a resource limit.¶
TODO: fill out currently missing registries:¶
MOQT version numbers¶
Setup parameters¶
Track Request parameters¶
Subscribe Error codes¶
Announce Error codes¶
Track format numbers¶
Message types¶
Object headers¶
TODO: register the URI scheme and the ALPN¶
TODO: the MOQT spec should establish the IANA registration table for MoQ Streaming Formats. Each MoQ streaming format can then register its type in that table. The MoQ Streaming Format type MUST be carried as the leading varint in catalog track objects.¶