Negotiating Media Multiplexing Using the Session Description Protocol (SDP)
EricssonHirsalantie 1102420JorvasFinlandchrister.holmberg@ericsson.comGoogleKungsbron 2Stockholm11122Swedenharald@alvestrand.noCisco400 3rd Avenue SW, Suite 350CalgaryABT2P 4H2Canadafluffy@iii.ca
Transport
MMUSIC Working GroupRTPSDPBundleMultiplexingRTCWEBCLUERTCWEBMMUSICAVTWEBBrowser
This specification defines a new Session Description
Protocol (SDP) Grouping Framework
extension, 'BUNDLE'. The extension can be used with the
SDP Offer/Answer mechanism to negotiate the usage of a
single address:port combination (BUNDLE address) for receiving media,
referred to as bundled media, specified by multiple
SDP media descriptions ("m=" lines).
To assist endpoints in negotiating the use of bundle this
specification defines a new SDP attribute, 'bundle-only',
which can be used to request that specific media is only
used if bundled. The specification also updates RFC 3264,
to allow usage of zero port values without meaning that
media is rejected.
There are multiple ways to correlate the bundled RTP
packets with the appropriate media descriptions. This
specification defines a new Real-time Transport Protocol
(RTP) source description (SDES) item and a new RTP header
extension that provides an additional way to do this
correlation by using them to carry a value that
associates the RTP/RTCP packets with a specific media
description.
When multimedia communications are established, each 5-tuple reserved for an individual media stream
consume additional resources (especially when Interactive Connectivity Establishment (ICE)
is used). For this reason, it is
attractive to use a 5-tuple for multiple media streams.
This specification defines a way to use a single address:port combination (BUNDLE address) for
receiving media specified by multiple SDP media descriptions ("m=" lines).
This specification defines a new SDP Grouping Framework extension called 'BUNDLE'. The extension can be used with the Session Description
Protocol (SDP) Offer/Answer mechanism
to negotiate the usage of a BUNDLE group. Within the BUNDLE group, a BUNDLE address is used
for receiving media specified by multiple "m=" lines. This is referred to as bundled media.
The offerer and answerer use
the BUNDLE extension to negotiate the BUNDLE addresses, one for the offerer (offerer BUNDLE
address) and one for the answerer (answerer BUNDLE address), to be used for receiving
the bundled media specified by a BUNDLE group. Once the offerer and the answerer have
negotiated a BUNDLE group, they associate their respective BUNDLE address with each "m=" line
in the BUNDLE group. The BUNDLE addresses are used to receive all media specified by
the BUNDLE group.
The use of a BUNDLE group and a BUNDLE address also allows the usage of a single set of
Interactive Connectivity Establishment (ICE)
candidates for multiple "m=" lines.
This specification also defines a new SDP attribute, 'bundle-only', which can be used to
request that specific media is only used if kept within a BUNDLE group. The specification
also updates RFC 3264, to allow usage of zero port values without meaning that media is rejected.
As defined in RFC 4566 , the
semantics of assigning the same transport address (IP address and port) to multiple
"m=" lines are undefined, and there is no grouping defined by such means. Instead, an
explicit grouping mechanism needs to be used to express the intended semantics. This
specification provides such an extension.
This specification also updates sections 5.1, 8.1 and 8.2 of RFC 3264 . The update allows an answerer to assign a non-zero port value to an "m=" line in an
SDP answer, even if the "m=" line in the associated SDP offer contained a zero port value.
This specification also defines a new Real-time Transport Protocol (RTP) source description (SDES) item, 'MID', and a new RTP SDES header
extension that can be used to associate RTP streams with media descriptions.
SDP bodies can contain multiple BUNDLE groups. A given BUNDLE address MUST only be associated
with a single BUNDLE group. The procedures in this specification apply independently to a given
BUNDLE group. All RTP based media flows described by a single BUNDLE group belong to a single
RTP session .
The BUNDLE extension is backward compatible. Endpoints that do not support the extension
are expected to generate offers and answers without an SDP 'group:BUNDLE' attribute, and
are expected to associate a unique address with each "m=" line within an offer and answer, according
to the procedures in and
"m=" line: SDP bodies contain one or more media descriptions. Each media
description is identified by an SDP "m=" line.
5-tuple: A collection of the following values: source address, source
port, destination address, destination port, and transport-layer
protocol.
Unique address: An IP address and port combination that is associated with
only one "m=" line in an offer or answer.
Shared address: An IP address and port combination that is associated with
multiple "m=" lines within an offer or answer.
Offerer BUNDLE-tag: The first identification-tag in a given
SDP 'group:BUNDLE' attribute identification-tag list in an offer.
Answerer BUNDLE-tag: The first identification-tag in a given
SDP 'group:BUNDLE' attribute identification-tag list in an answer.
Offerer BUNDLE address: Within a given BUNDLE group, an IP address and
port combination used by an offerer to receive all media specified
by each "m=" line within the BUNDLE group.
Answerer BUNDLE address: Within a given BUNDLE group, an IP address and
port combination used by an answerer to receive all media specified
by each "m=" line within the BUNDLE group.
BUNDLE group: A set of "m=" lines, created using an SDP Offer/Answer
exchange, which uses the same BUNDLE address for receiving media.
Bundled "m=" line: An "m=" line, whose identification-tag
is placed in an SDP 'group:BUNDLE' attribute identification-tag list
in an offer or answer.
Bundle-only "m=" line: A bundled "m=" line with an associated
SDP 'bundle-only' attribute.
Bundled media: All media specified by a given BUNDLE group.
Initial offer: The first offer, within an SDP session (e.g. a SIP dialog
when the Session Initiation Protocol (SIP) is used to carry SDP), in which
the offerer indicates that it wants to create a given BUNDLE group.
Subsequent offer: An offer which contains a BUNDLE group that
has been created as part of a previous offer/answer exchange.
Identification-tag: A unique token value that is used to identify an
"m=" line. The SDP 'mid' attribute , associated with an "m=" line, carries
an unique identification-tag. The session-level SDP 'group' attribute
carries a list
of identification-tags, identifying the "m=" lines associated with that
particular 'group' attribute.
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 BCP 14, RFC 2119 .
The mechanism in this specification only applies to the Session
Description Protocol (SDP) , when used together with the SDP offer/answer
mechanism .
Declarative usage of SDP is out of scope of this document, and is
thus undefined.
This section defines a new SDP Grouping Framework extension
, 'BUNDLE'. The
BUNDLE extension can be used with the SDP Offer/Answer mechanism to negotiate
the usage of a single address:port combination (BUNDLE address) for receiving bundled media.
A single address:port combination is also used for sending bundled media. The address:port
combination used for sending bundled media MAY be the same as the BUNDLE address, used to receive
bundled media, depending on whether symmetric RTP is used.
All media associated with a BUNDLE group MUST be transport using the same
transport-layer protocol (e.g., UDP or TCP).
The BUNDLE extension is indicated using an SDP 'group' attribute
with a "BUNDLE" semantics value . An identification-tag is associated with each bundled
"m=" line, and each identification-tag is listed in the SDP 'group:BUNDLE'
attribute identification-tag list. Each "m=" line whose identification-tag
is listed in the identification-tag list is associated with a given
BUNDLE group.
SDP bodies can contain multiple BUNDLE groups. Any given bundled "m="
line MUST NOT be associated with more than one BUNDLE group.
NOTE: The order of the "m=" lines listed in the SDP 'group:BUNDLE' attribute
identification-tag list does not have to be the same as the order in which
the "m=" lines occur in the SDP.
defines the
detailed SDP Offer/Answer procedures for the BUNDLE extension.
This section defines a new SDP media-level attribute , 'bundle-only'. 'bundle-only' is a property attribute
, and hence has no value.
In order to ensure that an answerer that does not support the BUNDLE extension always
rejects a bundled "m=" line, the offerer can assign a zero port value to the "m="
line. According to an answerer
will reject such "m=" line.
By associating an SDP 'bundle-only' attribute with such "m=" line, the offerer can
request that the answerer accepts the "m=" line if the answerer supports the Bundle
extension, and if the answerer keeps the "m=" line within the associated BUNDLE group.
NOTE: Once the offerer BUNDLE address has been selected, the offerer does
not need to include the 'bundle-only' attribute in subsequent offers. By
associating the offerer BUNDLE address with an "m=" line of a subsequent offer,
the offerer will ensure that the answerer will either keep the "m=" line within
the BUNDLE group, or the answerer will have to reject the "m=" line.
The usage of the 'bundle-only' attribute is only defined for a bundled "m=" line with
a zero port value, within an offer. Other usage is unspecified.
defines the detailed SDP
Offer/Answer procedures for the 'bundle-only' attribute.
This section describes restrictions associated with the usage of
SDP parameters within a BUNDLE group. It also describes, when parameter
and attribute values have been associated with each bundled "m=" line, how to
calculate a value for the whole BUNDLE group.
The "c=" line nettype value associated with a bundled "m=" line MUST be 'IN'.
The "c=" line addrtype value associated with a bundled "m=" line MUST be 'IP4' or
'IP6'. The same value MUST be associated with each "m=" line.
NOTE: Extensions to this specification can specify usage of the BUNDLE
mechanism for other nettype and addrtype values than the ones listed above.
An offerer and answerer MUST use the rules and restrictions defined
in for
associating the SDP bandwidth (b=) line with bundled "m=" lines.
This section describes the SDP Offer/Answer procedures for:
Negotiating and creating a BUNDLE group; and
Selecting the BUNDLE addresses (offerer BUNDLE address and answerer BUNDLE address); and
Adding an "m=" line to a BUNDLE group; and
Moving an "m=" line out of a BUNDLE group; and
Disabling an "m=" line within a BUNDLE group.
The generic rules and procedures defined in and
also apply to the BUNDLE extension. For example, if an offer is rejected
by the answerer, the previously negotiated SDP parameters and characteristics
(including those associated with a BUNDLE group) apply. Hence, if an offerer
generates an offer in which the offerer wants to create a BUNDLE group,
and the answerer rejects the offer, the BUNDLE group is not created.
The procedures in this section are independent of the media type or
"m=" line proto value represented by a bundled "m=" line.
defines additional
considerations for RTP based media.
defines
additional considerations for the usage of the SDP 'bundle-only' attribute.
defines additional
considerations for the usage of Interactive Connectivity Establishment (ICE)
mechanism.
SDP offers and answers can contain multiple BUNDLE groups. The procedures in this
section apply independently to a given BUNDLE group.
When an offerer or answerer associates SDP attributes with a bundled "m=" line
(including any bundle-only "m=" line) associated with a shared address, IDENTICAL
and TRANSPORT mux category SDP attributes
are associated with the "m=" line only if the "m=" line is also associated with
the offerer/answerer BUNDLE-tag. Otherwise the offerer/answerer MUST NOT associate
such SDP attributes with the "m=" line. The rule above does not apply to a bundled
"m=" line associated with a unique address.
NOTE: As bundled "m=" lines (including any bundle-only "m=" line) associated with
a shared address will share the same IDENTICAL and TRANSPORT mux category SDP attributes,
and attribute values, there is no need to associate such SDP attributes with each "m=" line.
The attributes and attribute values are implicitly applied to each "m=" line.
The semantics of some SDP attributes only apply to specific types of media.
For example, the semantics of the SDP 'rtcp-mux' and SDP 'rtcp-mux-only' attributes
only apply to "m=" lines describing RTP-based media. However, as described in
, there are cases where IDENTICAL and TRANSPORT
mux category SDP attributes are only associated with the "m=" line associated with
the BUNDLE-tag. That means that media-specific IDENTICAL and TRANSPORT mux category
attributes can be associated with an "m=" line associated with another type of media.
When an offerer generates an initial offer, in order to create a
BUNDLE group, it MUST:
Assign a unique address to each "m=" line within the offer,
following the procedures in , unless the media line is a 'bundle-only'
"m=" line (see below); and
Add an SDP 'group:BUNDLE' attribute to the offer; and
Place the identification-tag of each bundled "m=" line in the
SDP 'group:BUNDLE' attribute identification-tag list; and
Indicate which unique address the offerer suggests as the
offerer BUNDLE address [].
If the offerer wants to request that the answerer accepts a given bundled "m=" line only if
the answerer keeps the "m=" line within the BUNDLE group, the offerer MUST:
Associate an SDP 'bundle-only' attribute [] with the "m=" line; and
Assign a zero port value to the "m=" line.
NOTE: If the offerer assigns a zero port value to an "m=" line, but does not also associate an
SDP 'bundle-only' attribute with the "m=" line, it is an indication that the offerer wants
to disable the "m=" line [].
[] shows
an example of an initial offer.
In the offer, the address associated with the "m=" line associated with the
offerer BUNDLE-tag indicates the address that the offerer suggests as
the offerer BUNDLE address.
The "m=" line associated with the offerer BUNDLE-tag MUST NOT contain a zero
port value or an SDP 'bundle-only' attribute.
The example shows an initial SDP offer. The offer includes two
"m=" lines in the SDP, and suggests that both are included in
a BUNDLE group. The audio "m=" line is associated with the
offerer BUNDLE-tag (placed first in the SDP group:BUNDLE
attribute identificatoin-id list).
When an answerer generates an answer that contains a BUNDLE group, the following general
SDP grouping framework restrictions, defined in , also apply to the BUNDLE group:
The answerer MUST NOT include a BUNDLE group in the answer, unless the
offerer requested the BUNDLE group to be created in the corresponding offer; and
The answerer MUST NOT include an "m=" line within a BUNDLE group, unless
the offerer requested the "m=" line to be within that BUNDLE group in the
corresponding offer.
If the answer contains a BUNDLE group, the answerer MUST:
Select an Offerer BUNDLE Address []; and
Select an Answerer BUNDLE Address [];
The answerer is allowed to select a new Answerer BUNDLE address each time it
generates an answer to an offer.
If the answerer does not want to keep an "m=" line within a BUNDLE group, it MUST:
Move the "m=" line out of the BUNDLE group []; or
Reject the "m=" line [];
If the answerer keeps a bundle-only "m=" line within the BUNDLE group,
it follows the procedures (associates the answerer BUNDLE address with the "m="
line etc) for any other "m=" line kept within the BUNDLE group.
If the answerer does not want to keep a bundle-only "m=" line within the
BUNDLE group, it MUST reject the "m=" line [].
The answerer MUST NOT associate an SDP 'bundle-only' attribute with any
"m=" line in an answer.
NOTE: If a bundled "m=" line in an offer contains a zero port value, but the "m=" line does not contain
an SDP 'bundle-only' attribute, it is an indication that the offerer wants to disable the "m="
line [].
In an offer, the address (unique or shared) associated with the bundled "m=" line
associated with the offerer BUNDLE-tag indicates the address that
the offerer suggests as the offerer BUNDLE address
[]. The answerer
MUST check whether that "m=" line fulfils the following criteria:
The answerer will not move the "m=" line out of the BUNDLE group
[]; and
The answerer will not reject the "m=" line []; and
The "m=" line does not contain a zero port value.
If all of the criteria above are fulfilled, the answerer MUST select the address
associated with the "m=" line as the offerer BUNDLE address. In the answer, the
answerer BUNDLE-tag represents the "m=" line, and the
address associated with the "m=" line in the offer becomes the offerer BUNDLE address.
If one or more of the criteria are not fulfilled, the answerer MUST select the next
identification-tag in the identification-tag list, and perform the same criteria
check for the "m=" line associated with that identification-tag. If there are no
more identification-tags in the identification-tag list, the answerer MUST NOT
create the BUNDLE group. In addition, unless the answerer rejects the whole offer,
the answerer MUST apply the answerer procedures for moving an "m=" line out of a
BUNDLE group []
to each bundled "m=" line in the offer when creating the answer.
[] shows an
example of an offerer BUNDLE address selection.
When the answerer selects a BUNDLE address for itself, referred to as the answerer
BUNDLE address, it MUST associate that address with each bundled "m=" line within the
created BUNDLE group in the answer.
The answerer MUST NOT associate the answerer BUNDLE address with an "m=" line that is
not within the BUNDLE group, or to an "m=" line that is within another BUNDLE group.
[] shows an example of an
answerer BUNDLE address selection.
When an answerer wants to move an "m=" line out of a BUNDLE group,
it MUST first check the following criteria:
In the corresponding offer, the "m=" line is associated with a shared
address (e.g. a previously selected offerer BUNDLE address); or
In the corresponding offer, an SDP 'bundle-only' attribute is
associated with the "m=" line, and the "m=" line contains a
zero port value.
If either criteria above is fulfilled, the answerer MUST reject the "m=" line
[].
Otherwise, if in the corresponding offer the "m=" line is associated with
a unique address, the answerer MUST associate a unique address with the
"m=" line in the answer (the answerer does not reject the "m=" line).
In addition, in either case above, the answerer MUST NOT place the
identification-tag, associated with the moved "m=" line, in the SDP
'group' attribute identification-tag list associated with the BUNDLE
group.
When an answerer rejects an "m=" line, it MUST associate an address with a zero port
value with the "m=" line in the answer, according to the procedures in .
In addition, the answerer MUST NOT place the identification-tag, associated with
the rejected "m=" line, in the SDP 'group' attribute identification-tag list
associated with the BUNDLE group.
The example shows an SDP answer, based on the SDP offer in
[].
The answers acceppts both "m=" lines in the BUNDLE group.
When an offerer receives an answer, if the answer contains a BUNDLE group, the offerer
MUST check that any bundled "m=" line in the answer was indicated as bundled in the
corresponding offer. If there is no mismatch, the offerer MUST use the offerer BUNDLE address,
selected by the answerer [],
as the address for each bundled "m=" line.
NOTE: As the answerer might reject one or more bundled "m=" lines, or move a bundled
"m=" line out of a BUNDLE group, each bundled "m=" line in the offer might not be
indicated as bundled in the answer.
If the answer does not contain a BUNDLE group, the offerer MUST process the answer
as a normal answer.
When an offerer generates a subsequent offer, it MUST associate
the previously selected offerer BUNDLE address [] with each bundled "m=" line
(including any bundle-only "m=" line), except if:
The offerer suggests a new offerer BUNDLE address
[]; or
The offerer wants to add a bundled "m=" line to the BUNDLE group
[]; or
The offerer wants to move a bundled "m=" line out of the BUNDLE group
[]; or
The offerer wants to disable the bundled "m=" line [].
In addition, the offerer MUST select an offerer BUNDLE-tag
[]
associated with the previously selected offerer BUNDLE address, unless
the offerer suggests a new offerer BUNDLE address.
When an offerer generates an offer, in which it suggests a new
offerer BUNDLE address [], the offerer MUST:
Assign the address (shared address) to each "m=" line within the BUNDLE group; or
Assign the address (unique address) to one bundled "m=" line.
In addition, the offerer MUST indicate that the address is the
new suggested offerer BUNDLE address [].
NOTE: Unless the offerer associates the new suggested offerer BUNDLE address with each
bundled "m=" line, it can associate unique addresses with any number of bundled "m="
lines (and the previously selected offerer BUNDLE address to any remaining bundled
"m=" line) if it wants to suggest multiple alternatives for the new offerer
BUNDLE address.
When an offerer generates an offer, in which it wants to add a bundled "m=" line
to a BUNDLE group, the offerer MUST:
Assign a unique address to the added "m=" line; or
Assign the previously selected offerer BUNDLE address to the added "m=" line; or
If the offerer associates a new (shared address) suggested offerer BUNDLE address with each
bundled "m=" line [], also associate that address with the
added "m=" line.
In addition, the offerer MUST add the identification-tag associated with the added
"m=" line to the SDP 'group:BUNDLE' attribute identification-tag list with the
BUNDLE group [].
NOTE: Assigning a unique address to the "m=" line allows the answerer
to move the "m=" line out of the BUNDLE group [], without having to reject the "m=" line.
If the offerer associates a unique address with the added "m=" line, and if the offerer
suggests that address as the new offerer BUNDLE address [], the offerer BUNDLE-tag MUST represent
the added "m=" line [].
If the offerer associates a new suggested offerer BUNDLE address with each bundled "m=" line
[], including the added
"m=" line, the offerer BUNDLE-tag MAY represent the added "m=" line
[].
[]
shows an example where an offerer sends an offer in order to
add a bundled "m=" line to a BUNDLE group.
When an offerer generates an offer, in which it wants to move a bundled "m=" line out
of a BUNDLE group it was added to in a previous offer/answer transaction, the offerer:
MUST associate a unique address with the "m=" line; and
MUST NOT place the identification-tag associated with the "m=" line in
the SDP 'group:BUNDLE' attribute identification-tag list associated with
the BUNDLE group.
NOTE: If the removed "m=" line is associated with the previously selected
BUNDLE-tag, the offerer needs to suggest a new BUNDLE-tag
[].
NOTE: If an "m=" line, when being moved out of a BUNDLE group, is added to
another BUNDLE group, the offerer applies the procedures in
[] to
the "m=" line.
[]
shows an example of an offer for moving an "m=" line out of a BUNDLE group.
When an offerer generates an offer, in which it wants to disable a bundled "m=" line (added
to the BUNDLE group in a previous offer/answer transaction), the offerer:
MUST associate an address with a zero port value with the "m=" line,
following the procedures in ; and
MUST NOT place the identification-tag associated with the "m=" line in
the SDP 'group:BUNDLE' attribute identification-tag list associated with
the BUNDLE group.
[]
shows an example of an offer for disabling an "m=" line within a
BUNDLE group.
Each "m=" line within a BUNDLE group MUST use the same transport-
layer protocol. If bundled "m=" lines use different protocols on top
of the transport-layer protocol, there MUST exist a publicly
available specification which describes a mechanism, for this
particular protocol combination, how to associate received data
with the correct protocol.
In addition, if received data can be associated with more than
one bundled "m=" line, there MUST exist a publicly available
specification which describes a mechanism for associating the
received data with the correct "m=" line.
This document describes a mechanism to identify the
protocol of received data among the STUN, DTLS and SRTP protocols
(in any combination), when UDP is used as transport-layer protocol,
but does not describe how to identify different protocols transported on
DTLS. While the mechanism is generally applicable to other protocols and
transport-layer protocols, any such use requires further specification around
how to multiplex multiple protocols on a given transport-layer protocol,
and how to associate received data with the correct protocols.
Section 5.1.2 of describes a
mechanism to identify the protocol of a received packet among the STUN, Datagram
Transport Layer Security (DTLS) and SRTP protocols (in any combination).
If an offer or answer includes bundled "m=" lines that represent these protocols, the offerer
or answerer MUST support the mechanism described in , and no explicit negotiation is required in order to indicate support
and usage of the mechanism.
does not describe how to identify
different protocols transported on DTLS, only how to identify the DTLS protocol itself. If
multiple protocols are transported on DTLS, there MUST exist a specification describing a
mechanism for identifying each individual protocol. In addition, if a received DTLS packet
can be associated with more than one "m=" line, there MUST exist a specification which
describes a mechanism for associating the received DTLS packet with the correct "m=" line.
[] describes how to
associate the packets in a received SRTP stream with the correct "m=" line.
All RTP-based media within a single BUNDLE group belong to a
single RTP session .
Since a single RTP session is used for each bundle group, all
"m=" lines representing RTP-based media in a bundle group will
share a single SSRC numbering space .
The following rules and restrictions apply for a single RTP
session:
A specific payload type value can be used in multiple bundled "m=" lines
only if each codec associated with the payload type number shares an identical
codec configuration [].
The proto value in each bundled RTP-based "m=" line MUST be identical
(e.g. RTP/AVPF).
The RTP MID header extension MUST be enabled, by associating
an SDP 'extmap' attribute ,
with a 'urn:ietf:params:rtp- hdrext:sdes:mid' URI value, with each
bundled RTP-based "m=" line in every offer and answer.
A given SSRC MUST NOT transmit RTP packets using payload types that
originate from different bundled "m=" lines.
NOTE: The last bullet above is to avoid sending multiple media types
from the same SSRC. If transmission of multiple media types are done
with time overlap, RTP and RTCP fail to function. Even if done in
proper sequence this causes RTP Timestamp rate switching issues
. However,
once an SSRC has left the RTP session (by sending an RTCP BYE packet),
that SSRC can be reused by another source (possibly associated
with a different bundled "m=" line) after a delay of 5 RTCP reporting intervals
(the delay is to ensure the SSRC has timed out, in case the RTCP BYE
packet was lost ).
Multiple bundled "m=" lines might represent RTP based media. As all RTP based
media specified by a BUNDLE group belong to the same RTP session, in order
for a given payload type value to be used inside more than one bundled "m=" line,
all codecs associated with the payload type number MUST share an identical codec
configuration. This means that the codecs MUST share the same media type,
encoding name, clock rate and any parameter that can affect the codec configuration
and packetization. lists SDP attributes, whose attribute
values must be identical for all codecs that use the same payload type value.
NOTE: The text in this section is copied from Appendix B of JSEP. The
community has not yet agreed on the text.
As described in , RTP packets are associated with RTP
streams . Each RTP stream is identified by an SSRC
value, and each RTP packet includes an SSRC field that is
used to associate the packet with the correct RTP stream.
RTCP packets also use SSRCs to identify which RTP streams the
packet relates to. However, a RTCP packet can contain multiple SSRC
fields, in the course of providing feedback or reports on different RTP
streams, and therefore can be associated with multiple such streams.
In order to be able to process received RTP/RTCP packets
correctly, it must be possible to associate an RTP stream with
the correct "m=" line, as the "m=" line and SDP attributes
associated with the "m=" line contain information needed to
process the packets.
As all RTP streams associated with a BUNDLE group use the
same address:port combination for sending and receiving RTP/RTCP
packets, the local address:port combination cannot be used to
associate an RTP stream with the correct "m=" line. In addition,
multiple RTP streams might be associated with the same "m="
line.
An offerer and answerer can inform each other which SSRC
values they will use for an RTP stream by using the SDP 'ssrc'
attribute .
However, an offerer will not know which SSRC values the
answerer will use until the offerer has received the answer
providing that information. Due to this, before the offerer has
received the answer, the offerer will not be able to associate
an RTP stream with the correct "m=" line using the SSRC value
associated with the RTP stream. In addition, the offerer and
answerer may start using new SSRC values mid-session, without
informing each other using the SDP 'ssrc' attribute.
In order for an offerer and answerer to always be able to
associate an RTP stream with the correct "m=" line, the offerer
and answerer using the BUNDLE extension MUST support the
mechanism defined in , where the offerer and answerer
insert the identification-tag associated with an "m=" line
(provided by the remote peer) into RTP and RTCP packets associated
with a BUNDLE group.
When using this mechanism, the mapping from an SSRC to an
identification-tag is carried in RTP header extensions or RTCP SDES
packets, as specified in . Since a compound RTCP packet
can contain multiple RTCP SDES packets, and each RTCP SDES packet
can contain multiple chunks, a single RTCP packet can contain several
SSRC to identification-tag mappings. The offerer and answerer maintain
tables used for routing that are updated each time an RTP/RTCP packet
contains new information that affects how packets should be routed.
However, some implementations of may not include this identification-tag
in their RTP and RTCP traffic when using the BUNDLE mechanism,
and instead use a payload type based mechanism to associate RTP streams with SDP m= lines.
In this situation, each "m=" line MUST use unique payload type values,
in order for the payload type to be a reliable indicator of the relevant
"m=" line for the RTP stream.
Note that when using the payload type to associate RTP streams
with m= lines an RTP stream, identified by SSRC, will be mapped
to an "m=" line when the first packet of that RTP stream is received,
and the mapping will not be changed even if the payload type used by
that RTP stream changes. In other
words, the SSRC cannot to "move" to a different "m=" line
simply by changing the payload type.
Applications can implement RTP stacks in many different
ways. The algorithm below details one way that RTP streams can be
associated with m= lines, but is not meant to be prescriptive
about exactly how an RTP stack needs to be implemented.
Applications MAY use any algorithm that achieves equivalent results to
those described in the algorithm below.
To prepare to associate RTP streams with the correct "m="
line, the following steps MUST be followed for each BUNDLE group.
Construct a table mapping MID to "m=" line for each "m="
line in this BUNDLE group. Note that an "m=" line may only
have one MID.
Construct a table mapping SSRCs of incoming RTP streams to "m=" line for
each "m=" line in this BUNDLE group and for each SSRC
configured for receiving in that "m=" line.
Construct a table mapping the SSRC of each outgoing RTP stream to "m=line" for
each "m=" line in this BUNDLE group and for each SSRC
configured for sending in that "m=" line.
Construct a table mapping payload type to "m=" line for
each "m=" line in the BUNDLE group and for each payload type
configured for receiving in that "m=" line. If any payload
type is configured for receiving in more than one "m=" line
in the BUNDLE group, do not it include it in the table, as it
cannot be used to uniquely identify a "m=" line.
Note that for each of these tables, there can only be one
mapping for any given key (MID, SSRC, or PT). In other
words, the tables are not multimaps.
As "m=" lines are added or removed from the BUNDLE groups, or
their configurations are changed, the tables above MUST also be
updated.
When an RTP packet is received, it MUST be delivered to the RTP
stream corresponding to its SSRC. That RTP stream MUST then be
associated with the correct m= line within a BUNDLE group, for
additional processing, according to the following steps.
If the MID associated with the RTP stream is not in the
table mapping MID to “m=“ line, then the RTP stream is not
decoded and the payload data is discarded.
If the packet has a MID, and the packet's extended sequence number
is greater than that of the last MID update, as discussed in
, Section 4.2.6, update the MID associated
with the RTP stream to match the MID carried in the RTP packet, then
update the mapping tables to include an entry that maps the SSRC of
that RTP stream to the “m=“ line for that MID.
If the SSRC of the RTP stream is in the incoming SSRC
mapping table, check that the payload type used by the RTP
stream matches a payload type included on the matching
“m=“ line. If so, associate the RTP stream with that “m=“
line. Otherwise, the RTP stream is not decoded and the
payload data is discarded.
If the payload type used by the RTP stream is in the
payload type table, update the incoming SSRC mapping table
to include an entry that maps the RTP stream’s SSRC to the
“m=“ line for that payload type. Associate the RTP stream
with the corresponding “m=“ line.
Otherwise, mark the RTP stream as not for decoding and
discard the payload.
If the RTP packet contains one of more contributing source (CSRC)
identifiers, then each CSRC is looked up in the incoming SSRC table
and a copy of the RTP packet is associated with the corresponding
m= line for additional processing.
For each RTCP packet received (including each RTCP packet that
is part of a compound RTCP packet), the packet is processed as
usual by the RTP layer, then is passed to the “m=“ lines
corresponding to the RTP streams it contains information about
for additional processing.
This routing is type-dependent, as each kind of RTCP
packet has its own mechanism for associating it with the relevant
RTP streams.
RTCP packets for which no appropriate “m=“ line can be
identified MUST be processed as usual by the RTP layer,
updating the metadata associated with the corresponding RTP
streams, but are not passed to any “m=“ line. This situation
can occur with certain multiparty RTP topologies, or when
RTCP packets are sent containing a subset of the SDES
information.
Rules for additional processing of the various types of RTCP packets are
explained below.
If the RTCP packet is of type SDES, for each chunk in the packet
whose SSRC is found in the incoming SSRC table, deliver a copy
of the SDES packet to the "m=" line associated with that SSRC.
In addition, for any SDES MID items contained in these chunks,
if the MID is found in the table mapping MID to "m=" line,
update the incoming SSRC table to include an entry that
maps the RTP stream associated with chunk's SSRC to the "m=" line associated with that MID,
unless the packet is older than the packet that most recently
updated the mapping for this SSRC, as discussed in
, Section 4.2.6.
Note that if an SDES packet is received as part of a compound
RTCP packet, the SSRC to "m=" line mapping may not exist until
the SDES packet is handled (e.g., in the case where RTCP for a
source is received before any RTP packets). Therefore, when
processing a compound packet, any contained SDES packet MUST
be handled first. Note that this is a backwards change from
Section 6.1, which states that
"Each individual RTCP packet in the compound packet may be
processed independently with no requirements upon the order
or combination of packets".
If the RTCP packet is of type BYE, it indicates that the RTP streams
referenced in the packet are ending. Therefore, for each SSRC
indicated in the packet that is found in the incoming SSRC table,
first deliver a copy of the BYE packet to the "m=" line associated
with that SSRC, but then remove the entry for that SSRC from the
incoming SSRC table after an appropriate delay to
account for "straggler packets", as specified in , Section 6.2.1.
If the RTCP packet is of type SR or RR, for each report block in
the report whose "SSRC of source" is found in the outgoing
SSRC table, deliver a copy of the SR or RR packet to the "m=" line
associated with that SSRC. In addition, if the packet is of type
SR, and the sender SSRC for the packet is found in the
incoming SSRC table, deliver a copy of the SR packet to the "m=" line
associated with that SSRC.
If the implementation supports RTCP XR and the packet is of
type XR, as defined in ,
for each report block in the report whose "SSRC of source" is
is found in the outgoing SSRC table, deliver a copy of the
XR packet to the "m=" line associated with that SSRC.
In addition, if the sender SSRC for the packet is found in the
incoming SSRC table, deliver a copy of the XR packet to the "m=" line
associated with that SSRC.
If the RTCP packet is a feedback message of type RTPFB or PSFB,
as defined in , it will contain a
media source SSRC, and this SSRC is used for routing certain
subtypes of feedback messages. However, several subtypes of
PSFB messages include target SSRC(s) in a section called
Feedback Control Information (FCI). For these messages,
the target SSRC(s) are used for routing.
If the RTCP packet is a feedback packet that does not include
target SSRCs in its FCI section, and the media source SSRC is
found in the outgoing SSRC table, deliver the
feedback packet to the "m=" line associated with that SSRC.
RTPFB and PSFB types that are handled in this way include:
(PT=RTPFB, FMT=1).
(PT=PSFB, FMT=1).
(PT=PSFB, FMT=2).
(PT=PSFB, FMT=3).
If the RTCP packet is a feedback message that does include target
SSRC(s) in its FCI section, it can either be a request or a
notification. Requests reference a RTP stream that is being
sent by the message recipient, whereas notifications are responses
to an earlier request, and therefore reference a RTP stream that
is being received by the message recipient.
If the RTCP packet is a feedback request that includes target SSRC(s),
for each target SSRC that is found in the outgoing SSRC table,
deliver a copy of the RTCP packet to the "m=" line associated with
that SSRC. PSFB types that are handled in this way include:
(PT=PSFB, FMT=4).
(PT=PSFB, FMT=5).
(PT=PSFB, FMT=7).
(PT=PSFB, FMT=TBD).
If the RTCP packet is a feedback notification that include target SSRC(s),
for each target SSRC that is found in the incoming SSRC table,
deliver a copy of the RTCP packet to the "m=" line associated with
the RTP stream with matching SSRC. PSFB types that are handled in this way include:
(PT=PSFB, FMT=6). This message
is a notification in response to a prior TSTR.
If the RTCP packet is of type APP, then it is handled in an application
specific manner. If the application does not recognise the APP packet,
then it MUST be discarded.
Within a BUNDLE group, the offerer and answerer MUST enable
RTP/RTCP multiplexing for the RTP-based media specified by
the BUNDLE group.
When RTP/RTCP multiplexing is enabled, the same address:port
combination will be used for sending all RTP packets
and the RTCP packets associated with the BUNDLE group. Each endpoint
will send the packets towards the BUNDLE address of the other
endpoint. The same address:port combination MAY be used
for receiving RTP packets and RTCP packets.
This section describes how an offerer and answerer use the
SDP 'rtcp-mux' attribute and the SDP 'rtcp-mux-only' attribute
to negotiate usage of RTP/RTCP multiplexing for RTP-based media associated
with a BUNDLE group.
The mux category of the
SDP 'rtcp-mux' and 'rtcp-mux-only' attributes is IDENTICAL.
describes
the details regarding which bundled "m=" lines an offerer and answerer
associates the attributes with.
RTP/RTCP multiplexing only applies to RTP-based media. However, as described in
, within a BUNDLE
group the SDP 'rtcp-mux' and SDP 'rtcp-mux-only' attributes might be associated with
a non-RTP-based bundled "m=" line.
When an offerer generates an initial offer, if the offer contains
one or more RTP-based bundled "m=" lines (or, if there is a chance that RTP-based
"m=" lines will later be added to the BUNDLE group), the offerer MUST
associate an SDP 'rtcp-mux' attribute with one or more "m=" lines, following the procedures
for IDENTICAL mux category attributes in . In addition, the offerer MAY associate an
SDP 'rtcp-mux-only' attribute with the same "m=" lines.
NOTE: Whether the offerer associates the SDP 'rtcp-mux-only' attribute
depends on whether the offerer supports fallback to usage of a separate
port for RTCP in case the answerer moves one or more RTP-based "m=" line out of
the BUNDLE group in the answer.
NOTE: If the offerer associates an SDP 'rtcp-mux' attribute with one or
more bundled "m=" lines, but does not associate an SDP 'rtcp-mux-only' attribute,
the offerer can also associate an SDP 'rtcp' attribute with one or more RTP-based "m=" line in order to provide a
fallback port for RTCP, as described in . However, the fallback port will only be used for RTP-based
"m=" lines moved out of the BUNDLE group by the answerer.
In the initial offer, the address:port combination for RTCP MUST be unique in each
bundled RTP-based "m=" line (excluding a bundle-only "m=" line), similar to RTP.
When an answerer generates an answer, if the answerer supports RTP-based media,
and if a bundled "m=" line in the offer contained an SDP 'rtcp-mux' attribute,
the answerer MUST enable usage of RTP/RTCP multiplexing, even if there currently
are no RTP-based "m=" lines within the BUNDLE group. The answerer MUST associate
an SDP 'rtcp-mux' attribute with "m=" lines within the BUNDLE group in the answer
following the procedures for IDENTICAL mux category attributes in . In addition, if the "m=" line in the offer
contained an an SDP "rtcp-mux-only" attribute, the answerer MUST associate an SDP
"rtcp-mux-only" attribute with the "m=" line in the answer.
If the "m=" line associated with the offerer BUNDLE-tag in the offer contained an
SDP 'rtcp-mux-only' attribute, and if the answerer moves an RTP-based "m=" line
out of the BUNDLE group in the answer , the answerer MUST either associate the attribute with
the moved "m=" line (and enable RTP/RTCP multiplexing for the media associated with
the "m=" line), or reject the "m=" line .
The answerer MUST NOT associate an SDP 'rtcp' attribute with any "m=" line
within the BUNDLE group in the answer. The answerer will use the port value
of the selected offerer BUNDLE address for sending RTP and RTCP packets associated
with each RTP-based bundled "m=" line towards the offerer.
If the usage of RTP/RTCP multiplexing within a BUNDLE group has been
negotiated in a previous offer/answer transaction, the answerer MUST
associate an SDP 'rtcp-mux' attribute with the "m=" line associated with
the answerer BUNDLE-tag in the answer. It is not possible to disable
RTP/RTCP multiplexing within a BUNDLE group.
When an offerer receives an answer, if the answerer has accepted
the usage of RTP/RTCP multiplexing (see ),
the answerer follows the procedures for RTP/RTCP multiplexing defined
in . The
offerer will use the port value associated with the answerer
BUNDLE address for sending RTP and RTCP packets associated with each
RTP-based bundled "m=" line towards the answerer.
NOTE: It is considered a protocol error if the answerer has not
accepted the usage of RTP/RTCP multiplexing for RTP-based "m=" lines
that the answerer included in the BUNDLE group.
When an offerer generates a subsequent offer, the offerer MUST associate
an SDP 'rtcp-mux' attribute with a bundled "m=" line, following the procedures
for IDENTICAL mux category attributes in .
If the offerer wants to add a bundled RTP-based "m=" line
to the BUNDLE group, it MAY also associate an SDP 'rtcp-mux-only' attribute
with a bundled "m=", following the procedures for IDENTICAL mux category
attributes in .
This allows the offerer to mandate RTP/RTCP multiplexing for the added "m="
line (or the "m=" line to be rejected by the answerer) even if the answerer
does not accept the "m=" line within the BUNDLE group.
This section describes how to use the BUNDLE grouping extension together
with the Interactive Connectivity Establishment (ICE) mechanism .
The generic procedures for negotiating usage of ICE using SDP, defined
in , also apply to usage of ICE
with BUNDLE, with the following exceptions:
When BUNDLE addresses for a BUNDLE group have been selected for both endpoints, ICE
connectivity checks and keep-alives only need to be performed for the whole BUNDLE group,
instead of per bundled "m=" line.
Among bundled "m=" lines (including any bundle-only "m=" line) with which the offerer
has associated a shared address, the offerer only associates ICE-related media-level
SDP attributes with the "m=" line associated with the offerer BUNDLE-tag, following
the procedures in .
Among "m=" lines with which the answerer has associated a shared address within
a BUNDLE group, the answerer only associates ICE-related media-level SDP attributes
with the "m=" line associated with the answerer BUNDLE-tag, following the procedures in
.
Support and usage of ICE mechanism together with the BUNDLE extension is OPTIONAL.
When an offerer associates a unique address with a bundled "m=" line (excluding
any bundle-only "m=" line), the offerer MUST associate SDP 'candidate' attributes (and
other applicable ICE-related media-level SDP attributes), containing unique ICE properties
(candidates etc), with the "m=" line, according to the procedures in .
When an offerer associates a shared address with a bundled "m=" line, the offerer
MUST associate SDP 'candidate' attributes (and other applicable ICE-related
media-level SDP attributes) with the "m=" line following the procedures in
.
When an answerer associates a shared address with an "m=" line within a BUNDLE group, if the
answerer MUST associate SDP 'candidate' attributes (and other applicable ICE-related
media-level SDP attributes) with the "m=" line following the procedures in
.
NOTE: As most ICE-related media-level SDP attributes belong to the TRANSPORT mux category
, the offerer and answerer follow the
procedures in when
deciding whether to associate an attribute with a bundled "m=" line. However, in the
case of ICE-related media-level attributes, the rules apply to all attributes (see note below),
even if they belong to a different mux category.
NOTE: The following ICE-related media-level SDP attributes are defined in
: 'candidiate', 'remote-candidates', 'ice-mismatch',
'ice-ufrag', 'ice-pwd', and 'ice-pacing'.
When an offerer generates an initial offer, the offerer MUST associate ICE-related media-level
SDP attributes with bundled "m=" lines forllowin the procedures in [].
When an answerer generates an answer that contains a BUNDLE group,
the answer MUST associate ICE-related SDP attributes to "m=" lines
within the BUNDLE group according to [].
When an offerer receives an answer, if the answerer supports and uses the ICE
mechanism and the BUNDLE extension, the offerer MUST associate the ICE properties
associated with the offerer BUNDLE address, selected by the
answerer [],
with each bundled "m=" line.
When an offerer generates a subsequent offer, it MUST associate
ICE properties to bundled "m=" lines following the procedures in
[].
One or more media streams within a BUNDLE group might use
the Datagram Transport Layer Security (DTLS) protocol
in order to encrypt the data, or to negotiate encryption keys
if another encryption mechanism is used to encrypt media.
When DTLS is used within a BUNDLE group, the following rules
apply:
There can only be one DTLS association
associated with the BUNDLE group; and
Each usage of the DTLS association within the BUNDLE
group MUST use the same mechanism for determining
which endpoints (the offerer or answerer) become
DTLS client and DTLS server; and
Each usage of the DTLS association within the Bundle
group MUST use the same mechanism for determining
whether an offer or answer will trigger the
establishment of a new DTLS association, or whether
an existing DTLS association will be used; and
If the DTLS client supports DTLS-SRTP
it MUST include the 'use_srtp' extension
in the DTLS ClientHello message
,
The client MUST include the extension even if the usage
of DTLS-SRTP is not negotiated as part of the
multimedia session (e.g., SIP session .
NOTE: The inclusion of the 'use_srtp' extension during the initial
DTLS handshake ensures that a DTLS renegotiation will not be required
in order to include the extension, in case DTLS-SRTP encrypted media
is added to the BUNDLE group later during the multimedia session.
When RTP header extensions are used in the context of this
specification, the identifier used for a given extension MUST identify the same
extension across all the bundled media descriptions.
This section replaces the text of the following sections
of RFC 3264:
Section 5.1 (Unicast Streams).Section 8.2 (Removing a Media Stream).Section 8.4 (Putting a Unicast Media Stream on Hold).
For recvonly and sendrecv streams, the port number and address in the
offer indicate where the offerer would like to receive the media
stream. For sendonly RTP streams, the address and port number
indirectly indicate where the offerer wants to receive RTCP reports.
Unless there is an explicit indication otherwise, reports are sent to
the port number one higher than the number indicated. The IP address
and port present in the offer indicate nothing about the source IP
address and source port of RTP and RTCP packets that will be sent by
the offerer. A port number of zero in the offer indicates that the
stream is offered but MUST NOT be used. This has no useful semantics
in an initial offer, but is allowed for reasons of completeness,
since the answer can contain a zero port indicating a rejected stream
(Section 6). Furthermore, existing streams can be terminated by
setting the port to zero (Section 8). In general, a port number of
zero indicates that the media stream is not wanted.
For recvonly and sendrecv streams, the port number and address in the
offer indicate where the offerer would like to receive the media
stream. For sendonly RTP streams, the address and port number
indirectly indicate where the offerer wants to receive RTCP reports.
Unless there is an explicit indication otherwise, reports are sent to
the port number one higher than the number indicated. The IP address
and port present in the offer indicate nothing about the source IP
address and source port of RTP and RTCP packets that will be sent by
the offerer. A port number of zero in the offer by default indicates that the
stream is offered but MUST NOT be used, but an extension mechanism
might specify different semantics for the usage of a zero port value.
Furthermore, existing streams can be terminated by setting the port to
zero (Section 8). In general, a port number of zero by default indicates
that the media stream is not wanted.
A stream that is offered with a port of zero MUST be marked with port
zero in the answer. Like the offer, the answer MAY omit all
attributes present previously, and MAY list just a single media
format from amongst those in the offer.
A stream that is offered with a port of zero MUST by default be
marked with port zero in the answer, unless an extension mechanism,
which specifies semantics for the usage of a non-zero port value, is
used. If the stream is marked with port zero in the answer, the answer
MAY omit all attributes present previously, and MAY list
just a single media format from amongst those in the offer."
RFC 2543 [10] specified that placing a user on hold was accomplished
by setting the connection address to 0.0.0.0. Its usage for putting
a call on hold is no longer recommended, since it doesn't allow for
RTCP to be used with held streams, doesn't work with IPv6, and breaks
with connection oriented media. However, it can be useful in an
initial offer when the offerer knows it wants to use a particular set
of media streams and formats, but doesn't know the addresses and
ports at the time of the offer. Of course, when used, the port
number MUST NOT be zero, which would specify that the stream has been
disabled. An agent MUST be capable of receiving SDP with a
connection address of 0.0.0.0, in which case it means that neither
RTP nor RTCP should be sent to the peer.
RFC 2543 [10] specified that placing a user on hold was accomplished
by setting the connection address to 0.0.0.0. Its usage for putting
a call on hold is no longer recommended, since it doesn't allow for
RTCP to be used with held streams, doesn't work with IPv6, and breaks
with connection oriented media. However, it can be useful in an
initial offer when the offerer knows it wants to use a particular set
of media streams and formats, but doesn't know the addresses and
ports at the time of the offer. Of course, when used, the port
number MUST NOT be zero, if it would specify that the stream has been
disabled. However, an extension mechanism might specify different
semantics of the zero port number usage. An agent MUST be capable
of receiving SDP with a connection address of 0.0.0.0, in which case it
means that neither RTP nor RTCP should be sent to the peer.
SDP Offerers and Answerers
can associate identification-tags with "m=" lines within SDP Offers
and Answers, using the procedures in . Each identification-tag uniquely represents an "m=" line.
This section defines a new RTCP SDES item , 'MID', which is used to carry identification-tags within RTCP
SDES packets. This section also defines a new RTP SDES header extension
, which
is used to carry the 'MID' RTCP SDES item in RTP packets.
The SDES item and RTP SDES header extension make it possible for a receiver to associate
each RTP stream with with a specific "m=" line, with which the receiver has
associated an identification-tag, even if those "m=" lines are part of the same RTP session.
The RTP SDES header extension also ensures that the media recipient gets the identification-tag
upon receipt of the first decodable media and is able to associate the media with the
correct application.
A media recipient informs the media sender about the identification-tag
associated with an "m=" line through the use of an 'mid' attribute
. The media sender then
inserts the identification-tag in RTCP and RTP packets sent to the media recipient.
NOTE: This text above defines how identification-tags are carried in SDP Offers
and Answers. The usage of other signalling protocols for carrying identification-tags
is not prevented, but the usage of such protocols is outside the scope of this document.
defines general procedures
regarding the RTCP transmission interval. The RTCP MID SDES item SHOULD be sent in
the first few RTCP packets sent after joining the session, and SHOULD be sent regularly
thereafter. The exact number of RTCP packets in which this SDES item is sent is
intentionally not specified here, as it will depend on the expected packet loss
rate, the RTCP reporting interval, and the allowable overhead.
The RTP SDES header extension for carrying the 'MID' RTCP SDES SHOULD be included
in some RTP packets at the start of the session and whenever the SSRC changes. It might
also be useful to include the header extension in RTP packets that comprise access points in the media
(e.g., with video I-frames). The exact number of RTP packets in which this header
extension is sent is intentionally not specified here, as it will depend on expected
packet loss rate and loss patterns, the overhead the application can tolerate, and
the importance of immediate receipt of the identification-tag.
For robustness purpose, endpoints need to be prepared for situations where the
reception of the identification-tag is delayed, and SHOULD NOT terminate sessions
in such cases, as the identification-tag is likely to arrive soon.
The identification-tag payload is UTF-8 encoded, as in SDP.
The identification-tag is not zero terminated.
[RFC EDITOR NOTE: Please replace TBD with the assigned SDES
identifier value.]
The payload, containing the identification-tag, of the RTP SDES header extension element
can be encoded using either the one-byte or two-byte header . The identification-tag payload is UTF-8
encoded, as in SDP.
The identification-tag is not zero terminated. Note, that the set of header extensions
included in the packet needs to be padded to the next 32-bit boundary using zero
bytes .
As the identification-tag is included in either an RTCP SDES item or an RTP SDES header
extension, or both, there should be some consideration about the packet expansion
caused by the identification-tag. To avoid Maximum Transmission Unit (MTU) issues
for the RTP packets, the header extension's size needs to be taken into account when
encoding the media.
It is recommended that the identification-tag is kept short. Due to the properties of
the RTP header extension mechanism, when using the one-byte header, a tag that is 1-3 bytes
will result in a minimal number of 32-bit words used for the RTP SDES header extension,
in case no other header extensions are included at the same time. Note, do take into
account that some single characters when UTF-8 encoded will result in multiple octets.
The identification-tag MUST NOT contain any user information, and applications SHALL
avoid generating the identification-tag using a pattern that enables application
identification.
[RFC EDITOR NOTE: Please replace RFCXXXX with the RFC number
of this document.]
[RFC EDITOR NOTE: Please replace TBD with the assigned SDES
identifier value.]
This document adds the MID SDES item to the IANA "RTP SDES item
types" registry as follows:
[RFC EDITOR NOTE: Please replace RFCXXXX with the RFC number
of this document.]
This document defines a new extension URI in the RTP SDES Compact Header Extensions
sub-registry of the RTP Compact Header Extensions registry sub-registry, according
to the following data:
[RFC EDITOR NOTE: Please replace RFCXXXX with the RFC number
of this document.]
This document defines a new SDP media-level attribute,
'bundle-only', according to the following data:
[RFC EDITOR NOTE: Please replace RFCXXXX with the RFC number
of this document.]
This document registers the following semantics with IANA in the
"Semantics for the "group" SDP Attribute" subregistry (under the
"Session Description Protocol (SDP) Parameters" registry:
The security considerations defined in and apply to the BUNDLE extension. Bundle
does not change which information, e.g., RTP streams, flows over
the network, with the exception of the usage of the MID SDES item as
discussed below. Primarily it changes which addresses and ports, and
thus in which (RTP) sessions that the information is flowing in. This
affects the security contexts being used and can cause previously
separated information flows to share the same security context. This has very
little impact on the performance of the security mechanism of the RTP
sessions. In cases where one would have applied different security
policies on the different RTP streams being bundled, or where the
parties having access to the security contexts would have differed
between the RTP stream, additional analysis of the implications are
needed before selecting to apply BUNDLE.The identification-tag, independent of transport, RTCP SDES packet or
RTP header extension, can expose the value to parties beyond the
signaling chain. Therefore, the identification-tag values MUST be
generated in a fashion that does not leak user information, e.g.,
randomly or using a per-bundle group counter, and SHOULD be 3 bytes or
less, to allow them to efficiently fit into the MID RTP header
extension. Note that if implementations use different methods for
generating identification-tags this could enable fingerprinting of the
implementation making it vulnerable to targeted attacks. The
identification-tag is exposed on the RTP stream level when included in
the RTP header extensions, however what it reveals of the RTP media
stream structure of the endpoint and application was already possible to
deduce from the RTP streams without the MID SDES header extensions. As
the identification-tag is also used to route the media stream to the
right application functionality it is also important that the value
received is the one intended by the sender, thus integrity and the
authenticity of the source are important to prevent denial of service on
the application. Existing SRTP configurations and other security
mechanisms protecting the whole RTP/RTCP packets will provide the
necessary protection.When the BUNDLE extension is used, the set of configurations of the
security mechanism used in all the bundled media descriptions will need to
be compatible so that they can simultaneously used in parallel, at least
per direction or endpoint. When using SRTP this will be the case, at least
for the IETF defined key-management solutions due to their SDP attributes
(a=crypto, a=fingerprint, a=mikey) and their classification in .The security considerations of "RTP Header
Extension for the RTP Control Protocol (RTCP) Source Description Items"
requires that when RTCP is confidentiality protected that any SDES
RTP header extension carrying an SDES item, such as the MID RTP header
extension, is also protected using commensurate strength algorithms.
However, assuming the above requirements and recommendations are
followed there are no known significant security risks with leaving the
MID RTP header extension without confidentiality protection. Thus, the
requirements in RFC 7941 MAY be ignored for the MID RTP header
extension. Security mechanisms for RTP/RTCP are discussed in Options for
Securing RTP Sessions , for example SRTP can provide the necessary security
functions of ensuring the integrity and source authenticity.
The example below shows:
An offer, in which the offerer associates a unique address with each bundled "m=" line within the BUNDLE group.An answer, in which the answerer selects the offerer BUNDLE address, and then selects its own BUNDLE
address (the answerer BUNDLE address) and associates it with each bundled "m=" line within the BUNDLE group.
The example below shows:
An offer, in which the offerer associates a unique address with each bundled "m=" line within the BUNDLE group.An answer, in which the answerer rejects the offered BUNDLE group, and associates a unique address
with each "m=" line (following normal RFC 3264 procedures).
The example below shows:
A subsequent offer (the BUNDLE group has been created as part of a previous offer/answer
exchange), in which the offerer adds a new "m=" line, represented by the "zen"
identification-tag, to a previously negotiated BUNDLE group, associates a unique address
with the added "m=" line, and associates the previously selected offerer BUNDLE address with
each of the other bundled "m=" lines within the BUNDLE group.An answer, in which the answerer associates the answerer BUNDLE address with each
bundled "m=" line (including the newly added "m=" line) within the BUNDLE group.
The example below shows:
A subsequent offer (the BUNDLE group has been created as part of a previous offer/answer
transaction), in which the offerer moves a bundled "m=" line out of a BUNDLE group, associates a unique address with
the moved "m=" line, and associates the offerer BUNDLE address with each other bundled "m=" line within the BUNDLE group.An answer, in which the answerer moves the "m=" line out of the BUNDLE group, associates a unique address with
the moved "m=" line, and associates the answerer BUNDLE address with each of the remaining bundled "m=" line within the BUNDLE group.
The example below shows:
A subsequent offer (the BUNDLE group has been created as part of a previous offer/answer
transaction), in which the offerer disables a bundled "m=" line within a BUNDLE group, assigns a zero port number
to the disabled "m=" line, and associates the offerer BUNDLE address with each of the other bundled
"m=" lines within the BUNDLE group.An answer, in which the answerer moves the disabled "m=" line out of the BUNDLE group, assigns a zero port value
to the disabled "m=" line, and associates the answerer BUNDLE address with each of the remaining bundled "m=" line within
the BUNDLE group.
The usage of the SDP grouping extension for negotiating bundled media is
based on a similar alternatives proposed by Harald Alvestrand and Cullen
Jennings. The BUNDLE extension described in this document is based on
the different alternative proposals, and text (e.g., SDP examples)
have been borrowed (and, in some cases, modified) from those alternative
proposals.
The SDP examples are also modified versions from the ones in the Alvestrand
proposal.
Thanks to Paul Kyzivat, Martin Thomson, Flemming Andreasen, Thomas Stach,
Ari Keranen, Adam Roach, Christian Groves, Roman Shpount, Suhas Nandakumar, Nils
Ohlmeier, Jens Guballa, Raju Makaraju and Justin Uberti for
reading the text, and providing useful feedback.
Thanks to Bernard Aboba, Cullen Jennings, Peter Thatcher, Justin Uberti,
and Magnus Westerlund for providing the text for the section on
RTP/RTCP stream association.
Thanks to Magnus Westerlund, Colin Perkins and Jonathan Lennox for providing
help and text on the RTP/RTCP procedures.
Thanks to Spotify for providing music for the countless hours of
document editing.
[RFC EDITOR NOTE: Please remove this section when publishing]Changes from draft-ietf-mmusic-sdp-bundle-negotiation-38
Changes to RTP streaming mapping section based on text from Colin Perkins.The following GitHub pull requests were merged:https://github.com/cdh4u/draft-sdp-bundle/pull/34- Proposed updates to RTP processinghttps://github.com/cdh4u/draft-sdp-bundle/pull/35- fixed reference to receiver-id sectionChanges from draft-ietf-mmusic-sdp-bundle-negotiation-37
The following GitHub pull request was merged:https://github.com/cdh4u/draft-sdp-bundle/pull/33Changes from draft-ietf-mmusic-sdp-bundle-negotiation-36
The following GitHub pull requests were merged:https://github.com/cdh4u/draft-sdp-bundle/pull/32- extmap handling in BUNDLE.https://github.com/cdh4u/draft-sdp-bundle/pull/31- Additional Acknowledgement text added.https://github.com/cdh4u/draft-sdp-bundle/pull/30- MID SDES item security procedures updatedhttps://github.com/cdh4u/draft-sdp-bundle/pull/29- Appendix B of JSEP moved into BUNDLE.- Associating RTP/RTCP packets with SDP m- lines.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-35
Editorial changes on RTP streaming mapping section based on comments from Colin Perkins.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-34
RTP streams, instead of RTP packets, are associated with m- lines.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-33
Editorial changes based on comments from Eric Rescorla and Cullen Jennings:- Changes regarding usage of RTP/RTCP multiplexing attributes.- Additional text regarding associating RTP/RTCP packets with SDP m- lines.- Reference correction.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-32
Editorial changes based on comments from Eric Rescorla and Cullen Jennings:- Justification for mechanism added to Introduction.- Clarify that the order of m- lines in the group:BUNDLE
attribute does not have to be the same as the order in which
the m- lines are listed in the SDP.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-31
Editorial changes based on GitHub Pull requests by Martin Thomson:- https://github.com/cdh4u/draft-sdp-bundle/pull/2- https://github.com/cdh4u/draft-sdp-bundle/pull/1Editorial change based on comment from Diederick Huijbers (9th July 2016).Changes based on comments from Flemming Andreasen (21st June 2016):- Mux category for SDP bundle-only attribute added.- Mux category considerations editorial clarification.- Editorial changes.RTP SDES extension according to draft-ietf-avtext-sdes-hdr-ext.Note whether Design Considerations appendix is to be kept removed:- Appendix is kept within document.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-30
Indicating in the Abstract and Introduction that
the document updates RFC 3264.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-29
Change based on WGLC comment from Colin Perkins.- Clarify that SSRC can be reused by another source
after a delay of 5 RTCP reporting intervals.Change based on WGLC comment from Alissa Cooper.- IANA registry name fix.- Additional IANA registration information added.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-28
- Alignment with exclusive mux procedures.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-27
- Yet another terminology change.- Mux category considerations added.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-26
- ICE considerations modified: ICE-related SDP attributes only added
to the bundled m- line representing the selected BUNDLE address.- Reference to draft-ietf-mmusic-ice-sip-sdp added.- Reference to RFC 5245 replaced with reference to draft-ietf-ice-rfc5245bis.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-25
- RTP/RTCP mux procedures updated with exclusive RTP/RTCP mux considerations.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-24
- Reference and procedures associated with exclusive RTP/RTCP mux addedChanges from draft-ietf-mmusic-sdp-bundle-negotiation-23
- RTCP-MUX mandatory for bundled RTP m- lines- Editorial fixes based on comments from Flemming AndreasenChanges from draft-ietf-mmusic-sdp-bundle-negotiation-22
- Correction of Ari's family name- Editorial fixes based on comments from Thomas Stach- RTP/RTCP correction based on comment from Magnus Westerlund-- http://www.ietf.org/mail-archive/web/mmusic/current/msg14861.htmlChanges from draft-ietf-mmusic-sdp-bundle-negotiation-21
- Correct based on comment from Paul Kyzivat-- 'received packets' replaced with 'received data'Changes from draft-ietf-mmusic-sdp-bundle-negotiation-20
- Clarification based on comment from James Guballa- Clarification based on comment from Flemming AndreasenChanges from draft-ietf-mmusic-sdp-bundle-negotiation-19
- DTLS Considerations section added.- BUNDLE semantics added to the IANA Considerations- Changes based on WGLC comments from Adam Roach-- http://www.ietf.org/mail-archive/web/mmusic/current/msg14673.htmlChanges from draft-ietf-mmusic-sdp-bundle-negotiation-18
- Changes based on agreements at IETF#92-- BAS Offer removed, based on agreement at IETF#92.-- Procedures regarding usage of SDP "b=" line is replaced with a
reference to to draft-ietf-mmusic-sdp-mux-attributes.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-17
- Editorial changes based on comments from Magnus Westerlund.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-16
- Modification of RTP/RTCP multiplexing section, based
on comments from Magnus Westerlund.- Reference updates.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-15
- Editorial fix.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-14
- Editorial changes.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-13
Changes to allow a new suggested offerer BUNDLE address
to be assigned to each bundled m- line.Changes based on WGLC comments from Paul Kyzivat- Editorial fixesChanges from draft-ietf-mmusic-sdp-bundle-negotiation-12
Usage of SDP 'extmap' attribute addedSDP 'bundle-only' attribute scoped with "m=" lines with a zero port valueChanges based on WGLC comments from Thomas Stach- ICE candidates not assigned to bundle-only m- lines with a zero port value- Editorial changesChanges based on WGLC comments from Colin Perkins- Editorial changes:-- "RTP SDES item" -> "RTCP SDES item"-- "RTP MID SDES item" -> "RTCP MID SDES item"- Changes in section 10.1.1:-- "SHOULD NOT" -> "MUST NOT"-- Additional text added to the Note- Change to section 13.2:-- Clarify that mid value is not zero terminated- Change to section 13.3:-- Clarify that mid value is not zero terminated-- Clarify paddingChanges based on WGLC comments from Paul Kyzivat- Editorial changes:Changes based on WGLC comments from Jonathan Lennox- Editorial changes:- Defintion of SDP bundle-only attribute alligned with
structure in 4566bis draftChanges from draft-ietf-mmusic-sdp-bundle-negotiation-11
Editorial corrections based on comments from Harald Alvestrand.Editorial corrections based on comments from Cullen Jennings.Reference update (RFC 7160).Clarification about RTCP packet sending when RTP/RTCP multiplexing
is not used (http://www.ietf.org/mail-archive/web/mmusic/current/msg13765.html).Additional text added to the Security Considerations.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-10
SDP bundle-only attribute added to IANA Considerations.SDES item and RTP header extension added to Abstract and Introduction.Modification to text updating section 8.2 of RFC 3264.Reference corrections.Editorial corrections.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-09
Terminology change: "bundle-only attribute assigned to m= line" to
"bundle-only attribute associated with m= line".Editorial corrections.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-08
Editorial corrections.- "of"->"if" (8.3.2.5).- "optional"->"OPTIONAL" (9.1).- Syntax/ABNF for 'bundle-only' attribute added.- SDP Offer/Answer sections merged.- 'Request new offerer BUNDLE address' section addedChanges from draft-ietf-mmusic-sdp-bundle-negotiation-07
OPEN ISSUE regarding Receiver-ID closed.- RTP MID SDES Item.- RTP MID Header Extension.OPEN ISSUE regarding insertion of SDP 'rtcp' attribute in answers closed.- Indicating that, when rtcp-mux is used, the answerer MUST NOT include
an 'rtcp' attribute in the answer, based on the procedures in section 5.1.3 of
RFC 5761.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-06
Draft title changed.Added "SDP" to section names containing "Offer" or "Answer".Editorial fixes based on comments from Paul Kyzivat (http://www.ietf.org/mail-archive/web/mmusic/current/msg13314.html).Editorial fixed based on comments from Colin Perkins (http://www.ietf.org/mail-archive/web/mmusic/current/msg13318.html).- Removed text about extending BUNDLE to allow multiple RTP sessions within a BUNDLE group.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-05
Major re-structure of SDP Offer/Answer sections, to align with RFC 3264 structure.Additional definitions added.- Shared address.- Bundled "m=" line.- Bundle-only "m=" line.- Offerer suggested BUNDLE mid.- Answerer selected BUNDLE mid.Q6 Closed (IETF#88): An Offerer MUST NOT assign a shared address to multiple "m=" lines until it has
received an SDP Answer indicating support of the BUNDLE extension.Q8 Closed (IETF#88): An Offerer can, before it knows whether the Answerer supports the BUNDLE extension,
assign a zero port value to a 'bundle-only' "m=" line.SDP 'bundle-only' attribute section added.Connection data nettype/addrtype restrictions added.RFC 3264 update section added.Indicating that a specific payload type value can be used in multiple "m=" lines, if the value
represents the same codec configuration in each "m=" line.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-04
Updated Offerer procedures (http://www.ietf.org/mail-archive/web/mmusic/current/msg12293.html).Updated Answerer procedures (http://www.ietf.org/mail-archive/web/mmusic/current/msg12333.html).Usage of SDP 'bundle-only' attribute added.Reference to Trickle ICE document added.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-02
Mechanism modified, to be based on usage of SDP Offers
with both different and identical port number values, depending
on whether it is known if the remote endpoint supports the
extension.Cullen Jennings added as co-author.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-01
No changes. New version due to expiration.Changes from draft-ietf-mmusic-sdp-bundle-negotiation-00
No changes. New version due to expiration.Changes from draft-holmberg-mmusic-sdp-multiplex-negotiation-00
Draft name changed.Harald Alvestrand added as co-author."Multiplex" terminology changed to "bundle".Added text about single versus multiple RTP Sessions.Added reference to RFC 3550.
One of the main issues regarding the BUNDLE grouping extensions has been whether,
in SDP Offers and SDP Answers, the same port value should be inserted in "m="
lines associated with a BUNDLE group, as the purpose of the extension is to negotiate
the usage of a single address:port combination for media specified by the
"m=" lines. Issues with both approaches, discussed in the Appendix have been
raised. The outcome was to specify a mechanism which uses SDP Offers with both
different and identical port values.
Below are the primary issues that have been considered when defining the "BUNDLE"
grouping extension:
1) Interoperability with existing UAs.2) Interoperability with intermediary B2BUA- and proxy entities.3) Time to gather, and the number of, ICE candidates.4) Different error scenarios, and when they occur.5) SDP Offer/Answer impacts, including usage of port number value zero.
Consider the following SDP Offer/Answer exchange, where Alice sends an SDP Offer to Bob:
RFC 4961 specifies a way of doing symmetric RTP but that is an a later
invention to RTP and Bob can not assume that Alice supports RFC 4961. This
means that Alice may be sending RTP from a different port than 10000 or
10002 - some implementation simply send the RTP from an ephemeral
port. When Bob's endpoint receives an RTP packet, the only way that Bob
knows if it should be passed to the video or audio codec is by looking at
the port it was received on. This lead some SDP implementations to use the
fact that each "m=" line had a different port number to use that port
number as an index to find the correct m line in the SDP. As a result, some
implementations that do support symmetric RTP and ICE still use a SDP data
structure where SDP with "m=" lines with the same port such as:
will result in the second "m=" line being considered an SDP error
because it has the same port as the first line.
In an SDP Offer or SDP Answer, the media specified by an "m=" line can be
disabled/rejected by setting the port number value to zero. This is different
from e.g., using the SDP direction attributes, where RTCP traffic will
continue even if the SDP "inactive" attribute is indicated for the
associated "m=" line.
If each "m=" line associated with a BUNDLE group would contain different
port values, and one of those port values would be used for a BUNDLE address
associated with the BUNDLE group, problems would occur if an endpoint wants to
disable/reject the "m=" line associated with that port, by setting the port
value to zero. After that, no "m=" line would contain the port value which
is used for the BUNDLE address. In addition, it is unclear what would happen
to the ICE candidates associated with the "m=" line, as they are also used for
the BUNDLE address.
Some back to back user agents may be configured in a mode where if
the incoming call leg contains an SDP attribute the B2BUA does not
understand, the B2BUA still generates that SDP attribute in the Offer
for the outgoing call leg. Consider a B2BUA that did not understand
the SDP "rtcp" attribute, defined in RFC 3605, yet acted this way.
Further assume that the B2BUA was configured to tear down any call
where it did not see any RTCP for 5 minutes. In this case, if the B2BUA
received an Offer like:
It would be looking for RTCP on port 49172 but would not see any
because the RTCP would be on port 53020 and after five minutes, it would
tear down the call. Similarly, a B2BUA that did not understand BUNDLE yet
put BUNDLE in it's offer may be looking for media on the wrong port and
tear down the call. It is worth noting that a B2BUA that generated an
Offer with capabilities it does not understand is not compliant with the
specifications.
Sometimes intermediaries do not act as B2BUA, in the sense that
they don't modify SDP bodies, nor do they terminate SIP dialogs.
Still, however, they may use SDP information (e.g., IP address and
port) in order to control traffic gating functions, and to set
traffic policing rules. There might be rules which will trigger
a session to be terminated in case media is not sent or received
on the ports retrieved from the SDP. This typically occurs once the
session is already established and ongoing.
Sometimes intermediaries do not act as B2BUA, in the sense that
they don't modify SDP bodies, nor do they terminate SIP dialogs.
Still, however, they may use SDP information (e.g., codecs and
media types) in order to control bandwidth allocation functions.
The bandwidth allocation is done per "m=" line, which means that
it might not be enough if media specified by all "m=" lines
try to use that bandwidth. That may either simply lead to bad
user experience, or to termination of the call.
When using ICE, a candidate needs to be gathered for each port. This
takes approximately 20 ms extra for each extra "m=" line due to the NAT
pacing requirements. All of this gather can be overlapped with other
things while for exampe a web-page is loading to minimize the impact. If the client
only wants to generate TURN or STUN ICE candidates for one of the "m="
lines and then use trickle ICE
to get the non host ICE candidates for the rest of the "m=" lines, it MAY do
that and will not need any additional gathering time.
Some people have suggested a TURN extension to get a bunch of TURN
allocations at once. This would only provide a single STUN result so in
cases where the other end did not support BUNDLE, may cause more use of
the TURN server but would be quick in the cases where both sides
supported BUNDLE and would fall back to a successful call in the other
cases.