This page is not complete.
This article describes how the various WebRTC-related protocols interact with one another in order to create a connection and transfer data and/or media among peers.
Note: This page needs heavy rewriting for structural integrity and content completeness. Lots of info here is good but the organization is a mess since this is sort of a dumping ground right now.
Unfortunately, WebRTC can’t create connections without some sort of server in the middle. We call this the signal channel or signaling service. It’s any sort of channel of communication to exchange information before setting up a connection, whether by email, post card or a carrier pigeon... it’s up to you.
The information we need to exchange is the Offer and Answer which just contains the SDP mentioned below.
Peer A who will be the initiator of the connection, will create an Offer. They will then send this offer to Peer B using the chosen signal channel. Peer B will receive the Offer from the signal channel and create an Answer. They will then send this back to Peer A along the signal channel.
The configuration of an endpoint on a WebRTC connection is called a session description. The description includes information about the kind of media being sent, its format, the transfer protocol being used, the endpoint's IP address and port, and other information needed to describe a media transfer endpoint. This information is exchanged and stored using Session Description Protocol (SDP); if you want details on the format of SDP data, you can find it in RFC 2327.
When a user starts a WebRTC call to another user, a special description is created called an offer. This description includes all the information about the caller's proposed configuration for the call. The recipient then responds with an answer, which is a description of their end of the call. In this way, both devices share with one another the information needed in order to exchange media data. This exchange is handled using Interactive Connectivity Establishment (ICE, a protocol which lets two devices use an intermediary to exchange offers and answers even if the two devices are separated by Network Address Translation (NAT).
Each peer, then, keeps two descriptions on hand: the local description, describing itself, and the remote description, describing the other end of the call.
The offer/answer process is performed both when a call is first established, but also any time the call's format or other configuration needs to change. Regardless of whether it's a new call, or reconfiguring an existing one, these are the basic steps which must occur to exchange the offer and answer, leaving out the ICE layer for the moment:
RTCPeerConnection.createOffer()to create an offer.
RTCPeerConnection.setLocalDescription()to set that offer as the local description (that is, the description of the local end of the connection).
RTCPeerConnection.setRemoteDescription()to record it as the remote description (the description of the other end of the connection).
RTCPeerConnection.setLocalDescription(createdAnswer)to set the answer as its local description. The recipient now knows the configuration of both ends of the connection.
RTCPeerConnection.setRemoteDescription()to set the answer as the remote description for its end of the call. It now knows the configuration of both peers. Media begins to flow as configured.
Taking one step deeper into the process, we find that
remoteDescription, the properties which return these two descriptions, aren't as simple as they look. Because during renegotiation, an offer might be rejected because it proposes an incompatible format, it's necessary that each endpoint have the ability to propose a new format but not actually switch to it until it's accepted by the other peer. For that reason, WebRTC uses pending and current descriptions.
The current description (which is returned by the
RTCPeerConnection.currentRemoteDescription properties) represents the description currently in actual use by the connection. This is the most recent connection that both sides have fully agreed to use.
The pending description (returned by
RTCPeerConnection.pendingRemoteDescription) indicates a description which is currently under consideration following a call to
When reading the description (returned by
RTCPeerConnection.remoteDescription), the returned value is the value of
pendingRemoteDescription if there's a pending description (that is, the pending description isn't
null); otherwise, the current description (
currentRemoteDescription) is returned.
When changing the description by calling
setRemoteDescription(), the specified description is set as the pending description, and the WebRTC layer begins to evaluate whether or not it's acceptable. Once the proposed description has been agreed upon, the value of
currentRemoteDescription is changed to the pending description, and the pending description is set to null again, indicating that there isn't a pending description.
pendingLocalDescription contains not just the offer or answer under consideration, but any local ICE candidates which have already been gathered since the offer or answer was created. Similarly,
pendingRemoteDescription includes any remote ICE candidates which have been provided by calls to
See the individual articles on these properties and methods for more specifics.
As well as exchanging information about the media (discussed above in Offer/Answer and SDP), peers must exchange information about the network connection. This is known as an ICE candidate and details the available methods the peer is able to communicate (directly or through a TURN server). Typically, each peer will propose its best candidates first, making their way down the line toward their worse candidates. Ideally, candidates are UDP (since it's faster, and media streams are able to recover from interruptions relatively easily), but the ICE standard does allow TCP candidates as well.
Generally, ICE candidates using TCP are only going to be used when UDP is not available or is restricted in ways that make it not suitable for media streaming. Not all browsers support ICE over TCP, however.
ICE allows candidates to represent connections over either TCP or UDP, with UDP generally being preferred (and being more widely supported). Each protocol supports a few types of candidate, with the candidate types defining how the data makes its way from peer to peer.
UDP candidates (candidates with their
protocol set to
"udp") can be one of these types:
ipaddress is the actual, direct IP address of the remote peer.
"srflx"), but using TURN instead of STUN.
TCP candidates (that is, candidates whose
"tcp") can be of these types:
The ICE layer selects one of the two peers to serve as the controlling agent. This is the ICE agent which will make the final decision as to which candidate pair to use for the connection. The other peer is called the controlled agent. You can identify which one your end of the connection is by examining the value of
RTCIceCandidate.transport.role, although in general it doesn't matter which is which.
The controlling agent not only takes responsibility for making the final decision as to which candidate pair to use, but also for signaling that selection to the controlled agent by using STUN and an updated offer, if necessary. The controlled agent just waits to be told which candidate pair to use.
It's important to keep in mind that a single ICE session may result in the controlling agent choosing more than one candidate pair. Each time it does so and shares that information with the controlled agent, the two peers reconfigure their connection to use the new configuration described by the new candidate pair.
Once the ICE session is complete, the configuration that's currently in effect is the final one, unless an ICE reset occurs.
© 2005–2018 Mozilla Developer Network and individual contributors.
Licensed under the Creative Commons Attribution-ShareAlike License v2.5 or later.