Universitaet Bremen TZI

Postfach 330440 Bremen D-28359 Germany +49-421-218-63921 cabo@tzi.org

Zebra Technologies

820 W. Jackson Blvd. Suite 700 Chicago 60607 United States of America +1-847-634-6700 slemay@zebra.com

ARM Ltd.

110 Fulbourn Rd Cambridge CB1 9NJ Great Britain Hannes.tschofenig@gmx.net http://www.tschofenig.priv.at

Universitaet Bremen TZI

Postfach 330440 Bremen D-28359 Germany +49-421-218-63905 hartke@tzi.org

Tampere University of Technology

Korkeakoulunkatu 10 Tampere FI-33720 Finland bilhanan.silverajan@tut.fi

Microsoft

One Microsoft Way Redmond 98052 United States of America brian.raymor@microsoft.com

Applications Area (app) CORE The Constrained Application Protocol (CoAP), although inspired by HTTP, was designed to use UDP instead of TCP. The message layer of the CoAP over UDP protocol includes support for reliable delivery, simple congestion control, and flow control. Some environments benefit from the availability of CoAP carried over reliable transports such as TCP or TLS. This document outlines the changes required to use CoAP over TCP, TLS, and WebSockets transports.

The Constrained Application Protocol (CoAP) was designed for Internet of Things (IoT) deployments, assuming that UDP or DTLS over UDP can be used unimpeded. UDP is a good choice for transferring small amounts of data across networks that follow the IP architecture. Some CoAP deployments need to integrate well with existing enterprise infrastructures, where UDP-based protocols may not be well-received or may even be blocked by firewalls. Middleboxes that are unaware of CoAP usage for IoT can make the use of UDP brittle, resulting in lost or malformed packets. To address such environments, this document defines how to transport CoAP over TCP, CoAP over TLS, and CoAP over WebSockets. illustrates the layering:

Where NATs are present, CoAP over TCP can help with their traversal. NATs often calculate expiration timers based on the transport layer protocol being used by application protocols. Many NATs maintain TCP-based NAT bindings for longer periods based on the assumption that a transport layer protocol, such as TCP, offers additional information about the session life cycle. UDP, on the other hand, does not provide such information to a NAT and timeouts tend to be much shorter . Some environments may also benefit from the ability of TCP to exchange larger payloads, such as firmware images, without application layer segmentation and to utilize the more sophisticated congestion control capabilities provided by many TCP implementations. CoAP may be integrated into a Web environment where the front-end uses CoAP over UDP from IoT devices to a cloud infrastructure and then CoAP over TCP between the back-end services. A TCP-to-UDP gateway can be used at the cloud boundary to communicate with the UDP-based IoT device. To allow IoT devices to better communicate in these demanding environments, CoAP needs to support different transport protocols, namely TCP , in some situations secured by TLS . In addition, some corporate networks only allow Internet access via a HTTP proxy. In this case, the best transport for CoAP would be the WebSocket Protocol. The WebSocket protocol provides two-way communication between a client and a server after upgrading an HTTP/1.1 connection and may be available in an environment that blocks CoAP over UDP. Another scenario for CoAP over WebSockets is a CoAP application running inside a web browser without access to connectivity other than HTTP and WebSockets. This document specifies how to access resources using CoAP requests and responses over the TCP/TLS and WebSocket protocols. This allows connectivity-limited applications to obtain end-to-end CoAP connectivity either by communicating CoAP directly with a CoAP server accessible over a TCP/TLS or WebSocket connection or via a CoAP intermediary that proxies CoAP requests and responses between different transports, such as between WebSockets and UDP.

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 . This document assumes that readers are familiar with the terms and concepts that are used in and . A Block1 or Block2 option that includes an SZX value of 7. The payload of a CoAP message that is affected by a BERT Option in descriptive usage (Section 2.1 of ).

The request/response interaction model of CoAP over TCP is the same as CoAP over UDP. The primary differences are in the message layer. CoAP over UDP supports optional reliability by defining four types of messages: Confirmable, Non-confirmable, Acknowledgement, and Reset. TCP eliminates the need for the message layer to support reliability. As a result, message types are not defined in CoAP over TCP.

Conceptually, CoAP over TCP replaces most of the CoAP over UDP message layer with a framing mechanism on top of the byte stream provided by TCP/TLS, conveying the length information for each message that on datagram transports is provided by the UDP/DTLS datagram layer. TCP ensures reliable message transmission, so the CoAP over TCP messaging layer is not required to support acknowledgements or detection of duplicate messages. As a result, both the Type and Message ID fields are no longer required and are removed from the CoAP over TCP message format. All messages are also untyped. illustrates the difference between CoAP over UDP and CoAP over reliable transport. The removed Type (no type) and Message ID fields are indicated by dashes.

| +------------------->| | | | | | ACK [0xbc90] | | (-------) [------] | | 2.05 Content | | 2.05 Content | | (Token 0x71) | | (Token 0x71) | | "22.5 C" | | "22.5 C" | |<-------------------+ |<-------------------+ | | | | CoAP over UDP CoAP over reliable transport ]]>

A UDP-to-TCP gateway MUST discard all Empty messages (Code 0.00) after processing at the message layer. For Confirmable (CON), Non-Confirmable (NOM), and Acknowledgement (ACK) messages that are not Empty, their contents are repackaged into untyped messages.

The CoAP message format defined in , as shown in , relies on the datagram transport (UDP, or DTLS over UDP) for keeping the individual messages separate and for providing length information.

The CoAP over TCP message format is very similar to the format specified for CoAP over UDP. The differences are as follows: Since the underlying TCP connection provides retransmissions and deduplication, there is no need for the reliability mechanisms provided by CoAP over UDP. The “T” and “Message ID” fields in the CoAP message header are elided. The “Ver” field is elided as well. In constrast to the UDP message layer for UDP and DTLS, the CoAP over TCP message layer does not send a version number in each message. If required in the future, a new Capability and Settings Option (See ) could be defined to support version negotiation. In a stream oriented transport protocol such as TCP, a form of message delimitation is needed. For this purpose, CoAP over TCP introduces a length field with variable size. shows the adjusted CoAP message format with a modified structure for the fixed header (first 4 bytes of the CoAP over UDP header), which includes the length information of variable size, shown here as an 8-bit length.

4-bit unsigned integer. A value between 0 and 12 directly indicates the length of the message in bytes starting with the first bit of the Options field. Three values are reserved for special constructs: An 8-bit unsigned integer (Extended Length) follows the initial byte and indicates the length of options/payload minus 13. A 16-bit unsigned integer (Extended Length) in network byte order follows the initial byte and indicates the length of options/payload minus 269. A 32-bit unsigned integer (Extended Length) in network byte order follows the initial byte and indicates the length of options/payload minus 65805. The encoding of the Length field is modeled on CoAP Options (see section 3.1 of ). The following figures show the message format for the 0-bit, 16-bit, and the 32-bit variable length cases.

For example: A CoAP message just containing a 2.03 code with the token 7f and no options or payload would be encoded as shown in .

0x01 TKL = 1 ___/ Code = 2.03 --> 0x43 Token = 0x7f ]]>

The semantics of the other CoAP header fields are left unchanged.

CoAP requests and responses are exchanged asynchronously over the TCP/TLS connection. A CoAP client can send multiple requests without waiting for a response and the CoAP server can return responses in any order. Responses MUST be returned over the same connection as the originating request. Concurrent requests are differentiated by their Token, which is scoped locally to the connection. The connection is bi-directional, so requests can be sent both by the entity that established the connection and the remote host. Retransmission and deduplication of messages is provided by the TCP/TLS protocol. Since the TCP protocol provides ordered delivery of messages, the mechanism for reordering detection when observing resources is not needed. The value of the Observe Option in notifications MAY be empty on transmission and MUST be ignored on reception.

CoAP over WebSockets can be used in a number of configurations. The most basic configuration is a CoAP client retrieving or updating a CoAP resource located at a CoAP server that exposes a WebSocket endpoint (). The CoAP client acts as the WebSocket client, establishes a WebSocket connection, and sends a CoAP request, to which the CoAP server returns a CoAP response. The WebSocket connection can be used for any number of requests.

/ / \ CoAP | | Client \__/__/ <------------- \__\__/ Server | | | responses | | |___________| |___________| WebSocket =============> WebSocket Client Connection Server ]]>

The challenge with this configuration is how to identify a resource in the namespace of the CoAP server. When the WebSocket protocol is used by a dedicated client directly (i.e., not from a web page through a web browser), the client can connect to any WebSocket endpoint. This means it is necessary for the client to identify both the WebSocket endpoint (identified by a “ws” or “wss” URI) and the path and query of the CoAP resource within that endpoint from the same URI. When the WebSocket protocol is used from a web page, the choices are more limited , but the challenge persists. defines a new “coap+ws” URI scheme that identifies both a WebSocket endpoint and a resource within that endpoint as follows:

Another possible configuration is to set up a CoAP forward proxy at the WebSocket endpoint. Depending on what transports are available to the proxy, it could forward the request to a CoAP server with a CoAP UDP endpoint (), an SMS endpoint (a.k.a. mobile phone), or even another WebSocket endpoint. The client specifies the resource to be updated or retrieved in the Proxy-URI Option.

/ / \ CoAP / \ \ ---> / / \ CoAP | | Client \__/__/ <--- \__\__/ Proxy \__/__/ <--- \__\__/ Server | | | | | | | |___________| |___________| |___________| WebSocket ===> WebSocket UDP UDP Client Server Client Server ]]>

A third possible configuration is a CoAP server running inside a web browser (). The web browser initially connects to a WebSocket endpoint and is then reachable through the WebSocket server. When no connection exists, the CoAP server is unreachable. Because the WebSocket server is the only way to reach the CoAP server, the CoAP proxy should be a Reverse Proxy.

/ / \ CoAP / / \ ---> / \ \ CoAP | | Client \__/__/ <--- \__\__/ Proxy \__\__/ <--- \__/__/ Server | | | | | | | |___________| |___________| |___________| UDP UDP WebSocket <=== WebSocket Client Server Server Client ]]>

Further configurations are possible, including those where a WebSocket connection is established through an HTTP proxy. CoAP over WebSockets is intentionally very similar to CoAP over UDP. Therefore, instead of presenting CoAP over WebSockets as a new protocol, this document specifies it as a series of deltas from .

Before CoAP requests and responses are exchanged, a WebSocket connection is established as defined in Section 4 of . shows an example. The WebSocket client MUST include the subprotocol name “coap” in the list of protocols, which indicates support for the protocol defined in this document. Any later, incompatible versions of CoAP or CoAP over WebSockets will use a different subprotocol name. The WebSocket client includes the hostname of the WebSocket server in the Host header field of its handshake as per . The Host header field also indicates the default value of the Uri-Host Option in requests from the WebSocket client to the WebSocket server.

Once a WebSocket connection is established, CoAP requests and responses can be exchanged as WebSocket messages. Since CoAP uses a binary message format, the messages are transmitted in binary data frames as specified in Sections 5 and 6 of . The message format shown in is the same as the CoAP over TCP message format (see ) with one restriction. The Length (Len) field MUST be set to zero because the WebSockets frame contains the length.

The CoAP over TCP message format eliminates the Version field defined in CoAP over UDP. If CoAP version negotiation is required in the future, CoAP over WebSockets can address the requirement by the definition of a new subprotocol identifier that is negotiated during the opening handshake. Requests and response messages can be fragmented as specified in Section 5.4 of , though typically they are sent unfragmented as they tend to be small and fully buffered before transmission. The WebSocket protocol does not provide means for multiplexing. If it is not desirable for a large message to monopolize the connection, requests and responses can be transferred in a block-wise fashion as defined in . Empty messages (Code 0.00) MUST be ignored by the recipient (see also ).

CoAP requests and responses are exchanged asynchronously over the WebSocket connection. A CoAP client can send multiple requests without waiting for a response and the CoAP server can return responses in any order. Responses MUST be returned over the same connection as the originating request. Concurrent requests are differentiated by their Token, which is scoped locally to the connection. The connection is bi-directional, so requests can be sent both by the entity that established the connection and the remote host. Retransmission and deduplication of messages is provided by the WebSocket protocol. CoAP over WebSockets therefore does not make a distinction between Confirmable or Non-Confirmable messages, and does not provide Acknowledgement or Reset messages. Since the WebSocket protocol provides ordered delivery of messages, the mechanism for reordering detection when observing resources is not needed. The value of the Observe Option in notifications MAY be empty on transmission and MUST be ignored on reception.

When a client does not receive any response for some time after sending a CoAP request (or, similarly, when a client observes a resource and it does not receive any notification for some time), the connection between the WebSocket client and the WebSocket server may be lost or temporarily disrupted without the client being aware of it. To check the health of the WebSocket connection (and thereby of all active requests, if any), the client can send a Ping frame or an unsolicited Pong frame as specified in Section 5.5 of . There is no way to retransmit a request without creating a new one. Re-registering interest in a resource is permitted, but entirely unnecessary.

The WebSocket connection is closed as specified in Section 7 of . All requests for which the CoAP client has not received a response yet are cancelled when the connection is closed. If the client observes one or more resources over the WebSocket connection, then the CoAP server (or intermediary in the role of the CoAP server) MUST remove all entries associated with the client from the lists of observers when the connection is closed.

Signaling messages are introduced to allow peers to: Share characteristics such as maximum message size for the connection Shutdown the connection in an ordered fashion Terminate the connection in response to a serious error condition Signaling is a third basic kind of message in CoAP, after requests and responses. Signaling messages share a common structure with the existing CoAP messages. There is a code, a token, options, and an optional payload. (See Section 3 of for the overall structure, as adapted to the specific transport.)

A code in the 7.01-7.31 range indicates a Signaling message. Values in this range are assigned by the “CoAP Signaling Codes” sub-registry (see ). For each message, there is a sender and a peer receiving the message. Payloads in Signaling messages are diagnostic payloads (see Section 5.5.2 of ), unless otherwise defined by a Signaling message option.

Option numbers for Signaling messages are specific to the message code. They do not share the number space with CoAP options for request/response messages or with Signaling messages using other codes. Option numbers are assigned by the “CoAP Signaling Option Numbers” sub-registry (see ). Signaling options are elective or critical as defined in Section 5.4.1 of ). If a Signaling option is critical and not understood by the receiver, it MUST abort the connection (see ). If the option is understood but cannot be processed, the option documents the behavior.

Capability and Settings messages (CSM) are used for two purposes: Each capability option advertises one capability of the sender to the recipient. Setting options indicate a setting that will be applied by the sender. Most CSM Options are useful mainly as initial messages in the connection. Both capability and settings options are cumulative. A Capability and Settings message does not invalidate a previously sent capability indication or setting even if it is not repeated. A capability message without any option is a no-operation (and can be used as such). An option that is sent might override a previous value for the same option. The option defines how to handle this case if needed. Base values are listed below for CSM Options. These are the values for the Capability and Setting before any Capability and Settings messages sends a modified value. These are not default values for the option as defined in Section 5.4.4 in . A default value would mean that an empty Capability and Settings message would result in the option being set to its default value. Capability and Settings messages are indicated by the 7.01 code (CSM).

Number Applies to Name Format Length Base Value 1 CSM Server-Name string 1-255 (see below) A client can use the Server-Name critical option to indicate the default value for the Uri-Host Options in the messages that it sends to the server. It has the same restrictions as the Uri-Host Option (Section 5.10 of . For TLS, the initial value for the Server-Name Option is given by the SNI value. For Websockets, the initial value for the Server-Name Option is given by the HTTP Host header field.

The sender can use the Max-Message-Size elective option to indicate the maximum message size in bytes that it can receive. Number Applies to Name Format Length Base Value 2 CSM Max-Message-Size uint 0-4 1152 As per Section 4.6 of , the base value (and the value used when this option is not implemented) is 1152. A peer that relies on this option being indicated with a certain minimum value will enjoy limited interoperability.

Number Applies to Name Format Length Base Value 4 CSM Block-wise Transfer empty 0 (none) A sender can use the Block-wise Transfer elective Option to indicate that it supports the block-wise transfer protocol . If the option is not given, the peer has no information about whether block-wise transfers are supported by the sender or not. An implementation that supports block-wise transfers SHOULD indicate the Block-wise Transfer Option. If a Max-Message-Size Option is indicated with a value that is greater than 1152 (in the same or a different CSM message), the Block-wise Transfer Option also indicates support for BERT (see ).

In CoAP over TCP, Empty messages (Code 0.00) can always be sent and MUST be ignored by the recipient (see also ). This provides a basic keep-alive function that can refresh NAT bindings. In contrast, Ping and Pong messages are a bidirectional exchange. Upon receipt of a Ping message, a single Pong message is returned with the identical token. As with all Signaling messages, the recipient of a Ping or Pong message MUST ignore elective options it does not understand. Ping and Pong messages are indicated by the 7.02 code (Ping) and the 7.03 code (Pong).

Number Applies to Name Format Length Base Value 2 Ping, Pong Custody empty 0 (none) A peer replying to a Ping message can add a Custody elective option to the Pong message it returns. This option indicates that the application has processed all request/response messages that it has received in the present connection ahead of the Ping message that prompted the Pong message. (Note that there is no definition of specific application semantics of “processed”, but there is an expectation that the sender of the Ping leading to the Pong with a Custody Option should be able to free buffers based on this indication.) A Custody elective option can also be sent in a Ping message to explicitly request the return of a Custody Option in the Pong message. A peer is always free to indicate that it has finished processing all previous request/response messages by sending a Custody Option in a Pong message. A peer is also free NOT to send a Custody Option in case it is still processing previous request/response messages; however, it SHOULD delay its response to a Ping with a Custody Option until it can also return one.

A release message indicates that the sender does not want to continue maintaining the connection and opts for an orderly shutdown; the details are in the options. A diagnostic payload MAY be included. A release message will normally be replied to by the peer by closing the TCP/TLS connection. Messages may be in flight when the sender decides to send a Release message. The general expectation is that these will still be processed. Release messages are indicated by the 7.04 code (Release). Release messages can indicate one or more reasons using elective options. The following options are defined: Number Applies to Name Format Length Base Value 2 Release Bad-Server-Name empty 0 (none) The Bad-Server-Name elective option indicates that the default indicated by the CSM Server-Name Option is unlikely to be useful for this server. Number Applies to Name Format Length Base Value 4 Release Alternate-Address string 1-255 (none) The Alternative-Address elective option requests the peer to instead open a connection of the same kind as the present connection to the alternative transport address given. Its value is in the form “authority” as defined in Section 3.2 of . Number Applies to Name Format Length Base Value 6 Release Hold-Off uint 0-3 (none) The Hold-Off elective option indicates that the server is requesting that the peer not reconnect to it for the number of seconds given in the value.

An abort message indicates that the sender is unable to continue maintaining the connection and cannot even wait for an orderly release. The sender shuts down the connection immediately after the abort (and may or may not wait for a release or abort message or connection shutdown in the inverse direction). A diagnostic payload SHOULD be included in the Abort message. Messages may be in flight when the sender decides to send an abort message. The general expectation is that these will NOT be processed. Abort messages are indicated by the 7.05 code (Abort). Abort messages can indicate one or more reasons using elective options. The following option is defined: Number Applies to Name Format Length Base Value 2 Abort Bad-CSM-Option uint 0-2 (none) The Bad-CSM-Option Option indicates that the sender is unable to process the CSM option identified by its option number, e.g. when it is critical and the option number is unknown by the sender, or when there is parameter problem with the value of an elective option. More detailed information SHOULD be included as a diagnostic payload. One reason for an sender to generate an abort message is a general syntax error in the byte stream received. No specific option has been defined for this, as the details of that syntax error are best left to a diagnostic payload.

An encoded example of a Ping message with a non-empty token is shown in .

0x01 TKL = 1 ___/ Code = 7.02 Ping --> 0xe2 Token = 0x42 ]]>

An encoded example of the corresponding Pong message is shown in .

0x01 TKL = 1 ___/ Code = 7.03 Pong --> 0xe3 Token = 0x42 ]]>

The message size restrictions defined in Section 4.6 of CoAP to avoid IP fragmentation are not necessary when CoAP is used over a reliable byte stream transport. While this suggests that the Block-wise transfer protocol is also no longer needed, it remains applicable for a number of cases: large messages, such as firmware downloads, may cause undesired head-of-line blocking when a single TCP connection is used a UDP-to-TCP gateway may simply not have the context to convert a message with a Block Option into the equivalent exchange without any use of a Block Option (it would need to convert the entire blockwise exchange from start to end into a single exchange) The ‘Block-wise Extension for Reliable Transport (BERT)’ extends the Block protocol to enable the use of larger messages over a reliable transport. The use of this new extension is signaled by sending Block1 or Block2 Options with SZX == 7 (a “BERT option”). SZX == 7 is a reserved value in . In control usage, a BERT option is interpreted in the same way as the equivalent Option with SZX == 6, except that it also indicates the capability to process BERT blocks. As with the basic Block protocol, the recipient of a CoAP request with a BERT option in control usage is allowed to respond with a different SZX value, e.g. to send a non-BERT block instead. In descriptive usage, a BERT Option is interpreted in the same way as the equivalent Option with SZX == 6, except that the payload is allowed to contain a multiple of 1024 bytes (non-final BERT block) or more than 1024 bytes (final BERT block). The recipient of a non-final BERT block (M=1) conceptually partitions the payload into a sequence of 1024-byte blocks and acts exactly as if it had received this sequence in conjunction with block numbers starting at, and sequentially increasing from, the block number given in the Block Option. In other words, the entire BERT block is positioned at the byte position that results from multiplying the block number with 1024. The position of further blocks to be transferred is indicated by incrementing the block number by the number of elements in this sequence (i.e., the size of the payload divided by 1024 bytes). As with SZX == 6, the recipient of a final BERT block (M=0) simply appends the payload at the byte position that is indicated by the block number multiplied with 1024. The following examples illustrate BERT options. A value of SZX == 7 is labeled as “BERT” or as “BERT(nnn)” to indicate a payload of size nnn. In all these examples, a Block Option is decomposed to indicate the kind of Block Option (1 or 2) followed by a colon, the block number (NUM), more bit (M), and block size exponent (2**(SZX+4)) separated by slashes. E.g., a Block2 Option value of 33 would be shown as 2:2/0/32), or a Block1 Option value of 59 would be shown as 1:3/1/128.

shows a GET request with a response that is split into three BERT blocks. The first response contains 3072 bytes of payload; the second, 5120; and the third, 4711. Note how the block number increments to move the position inside the response body forward.

| | | | <------ 2.05 Content, 2:0/1/BERT(3072) | | | | GET, /status, 2:3/0/BERT ------> | | | | <------ 2.05 Content, 2:3/1/BERT(5120) | | | | GET, /status, 2:8/0/BERT ------> | | | | <------ 2.05 Content, 2:8/0/BERT(4711) | ]]>

demonstrates a PUT exchange with BERT blocks.

| | | | <------ 2.31 Continue, 1:0/1/BERT | | | | PUT, /options, 1:8/1/BERT(16384) ------> | | | | <------ 2.31 Continue, 1:8/1/BERT | | | | PUT, /options, 1:24/0/BERT(5683) ------> | | | | <------ 2.04 Changed, 1:24/0/BERT | | | ]]>

CoAP over UDP defines the “coap” and “coaps” URI schemes for identifying CoAP resources and providing a means of locating the resource.

CoAP over TCP uses the “coap+tcp” URI scheme. CoAP over TLS uses the “coaps+tcp” scheme. The rules from Section 6 of apply to both of these URI schemes. , Section 8 (Multicast CoAP) is not applicable to these schemes. Resources made available via one of the “coap+tcp” or “coaps+tcp” schemes have no shared identity with the other scheme or with the “coap” or “coaps” scheme, even if their resource identifiers indicate the same authority (the same host listening to the same port). The schemes constitute distinct namespaces and, in combination with the authority, are considered to be distinct origin servers.

The semantics defined in , Section 6.1, apply to this URI scheme, with the following changes: The port subcomponent indicates the TCP port at which the CoAP server is located. (If it is empty or not given, then the default port 5683 is assumed, as with UDP.)

The semantics defined in , Section 6.2, apply to this URI scheme, with the following changes: The port subcomponent indicates the TCP port at which the TLS server for the CoAP server is located. If it is empty or not given, then the default port 443 is assumed (this is different from the default port for “coaps”, i.e., CoAP over DTLS over UDP). When CoAP is exchanged over TLS port 443, the “TLS Application Layer Protocol Negotiation Extension” MUST be used to allow demultiplexing at the server-side.

For the first configuration discussed in , this document defines two new URIs schemes that can be used for identifying CoAP resources and providing a means of locating these resources: “coap+ws” and “coap+wss”. Similar to the “coap” and “coaps” schemes, the “coap+ws” and “coap+wss” schemes organize resources hierarchically under a CoAP origin server. The key difference is that the server is potentially reachable on a WebSocket endpoint instead of a UDP endpoint. The WebSocket endpoint is identified by a “ws” or “wss” URI that is composed of the authority part of the “coap+ws” or “coap+wss” URI, respectively, and the well-known path “/.well-known/coap” . The path and query parts of a “coap+ws” or “coap+wss” URI identify a resource within the specified endpoint which can be operated on by the methods defined by CoAP. The syntax of the “coap+ws” and “coap+wss” URI schemes is specified below in Augmented Backus-Naur Form (ABNF) . The definitions of “host”, “port”, “path-abempty” and “query” are the same as in .

The port component is OPTIONAL; the default for “coap+ws” is port 80, while the default for “coap+wss” is port 443. Fragment identifiers are not part of the request URI and thus MUST NOT be transmitted in a WebSocket handshake or in the URI options of a CoAP request.

The steps for decomposing a “coap+ws” or “coap+wss” URI into CoAP options are the same as specified in Section 6.4 of with the following changes: The <scheme> component MUST be “coap+ws” or “coap+wss” when converted to ASCII lowercase. A Uri-Host Option MUST only be included in a request when the <host> component does not equal the uri-host component in the Host header field in the WebSocket handshake. A Uri-Port Option MUST only be included in a request if |port| does not equal the port component in the Host header field in the WebSocket handshake. The steps to construct a URI from a request’s options are changed accordingly.

The security considerations of apply. Implementations of CoAP MUST use TLS version 1.2 or higher for CoAP over TLS. The general TLS usage guidance in SHOULD be followed. Guidelines for use of cipher suites and TLS extensions can be found in . TLS does not protect the TCP header. This may, for example, allow an on-path adversary to terminate a TCP connection prematurely by spoofing a TCP reset message. CoAP over WebSockets and CoAP over TLS-secured WebSockets do not introduce additional security issues beyond CoAP and DTLS-secured CoAP respectively . The security considerations of apply.

The guidance given by an Alternative-Address Option cannot be followed blindly. In particular, a peer MUST NOT assume that a successful connection to the Alternative-Address inherits all the security properties of the current connection. SNI vs. Server-Name: Any security negotiated in the TLS handshake is for the SNI name exchanged in the TLS handshake and checked against the certificate provided by the server. The Server-Name Option cannot be used to extend these security properties to the additional server name.

IANA is requested to create a third sub-registry for values of the Code field in the CoAP header (Section 12.1 of ). The name of this sub-registry is “CoAP Signaling Codes”. Each entry in the sub-registry must include the Signaling Code in the range 7.01-7.31, its name, and a reference to its documentation. Initial entries in this sub-registry are as follows: Code Name Reference 7.01 CSM [RFCthis] 7.02 Ping [RFCthis] 7.03 Pong [RFCthis] 7.04 Release [RFCthis] 7.05 Abort [RFCthis] All other Signaling Codes are Unassigned. The IANA policy for future additions to this sub-registry is “IETF Review or IESG Approval” as described in .

IANA is requested to create a sub-registry for signaling options similar to the CoAP Option Numbers Registry (Section 12.2 of ), with the single change that a fourth column is added to the sub-registry that is one of the codes in the Signaling Codes subregistry (). The name of this sub-registry is “CoAP Signaling Option Numbers”. Initial entries in this sub-registry are as follows: Number Applies to Name Reference 1 CSM Server-Name [RFCthis] 2 CSM Max-Message-Size [RFCthis] 4 CSM Block-wise-Transfer [RFCthis] 2 Ping, Pong Custody [RFCthis] 2 Release Bad-Server-Name [RFCthis] 4 Release Alternative-Address [RFCthis] 6 Release Hold-Off [RFCthis] 2 Abort Bad-CSM-Option [RFCthis] The IANA policy for future additions to this sub-registry is based on number ranges for the option numbers, analogous to the policy defined in Section 12.2 of .

IANA is requested to assign the port number 5683 and the service name “coap+tcp”, in accordance with . coap+tcp tcp IESG <iesg@ietf.org> IETF Chair <chair@ietf.org> Constrained Application Protocol (CoAP) [RFCthis] 5683 Similarly, IANA is requested to assign the service name “coaps+tcp”, in accordance with . However, no separate port number is used for “coaps” over TCP; instead, the ALPN protocol ID defined in is used over port 443. coaps+tcp tcp IESG <iesg@ietf.org> IETF Chair <chair@ietf.org> Constrained Application Protocol (CoAP) , [RFCthis] 443 (see also of [RFCthis]})

This document registers two new URI schemes, namely “coap+tcp” and “coaps+tcp”, for the use of CoAP over TCP and for CoAP over TLS over TCP, respectively. The “coap+tcp” and “coaps+tcp” URI schemes can thus be compared to the “http” and “https” URI schemes. The syntax of the “coap” and “coaps” URI schemes is specified in Section 6 of and the present document re-uses their semantics for “coap+tcp” and “coaps+tcp”, respectively, with the exception that TCP, or TLS over TCP is used as a transport protocol. IANA is requested to add these new URI schemes to the registry established with .

This document requests the registration of the Uniform Resource Identifier (URI) scheme “coap+ws”. The registration request complies with . coap+ws Permanent Defined in Section N of [RFCthis] The “coap+ws” URI scheme provides a way to identify resources that are potentially accessible over the Constrained Application Protocol (CoAP) using the WebSocket protocol. The scheme encoding conforms to the encoding rules established for URIs in , i.e., internationalized and reserved characters are expressed using UTF-8-based percent-encoding. The scheme is used by CoAP endpoints to access CoAP resources using the WebSocket protocol. None. See Section N of [RFCthis] IETF chair <chair@ietf.org> IESG <iesg@ietf.org> [RFCthis]

This document requests the registration of the Uniform Resource Identifier (URI) scheme “coap+wss”. The registration request complies with . coap+wss Permanent Defined in Section N of [RFCthis] The “coap+ws” URI scheme provides a way to identify resources that are potentially accessible over the Constrained Application Protocol (CoAP) using the WebSocket protocol secured with Transport Layer Security (TLS). The scheme encoding conforms to the encoding rules established for URIs in , i.e., internationalized and reserved characters are expressed using UTF-8-based percent-encoding. The scheme is used by CoAP endpoints to access CoAP resources using the WebSocket protocol secured with TLS. None. See Section N of [RFCthis] IETF chair <chair@ietf.org> IESG <iesg@ietf.org> [RFCthis]

IANA is requested to register the ‘coap’ well-known URI in the “Well-Known URIs” registry. This registration request complies with : coap IETF [RFCthis] None.

IANA is requested to assign the following value in the registry “Application Layer Protocol Negotiation (ALPN) Protocol IDs” created by : CoAP 0x63 0x6f 0x61 0x70 (“coap”) [RFCthis]

IANA is requested to register the WebSocket CoAP subprotocol under the “WebSocket Subprotocol Name Registry”: coap Constrained Application Protocol (CoAP) [RFCthis]

In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements. A Uniform Resource Identifier (URI) is a compact sequence of characters that identifies an abstract or physical resource. This specification defines the generic URI syntax and a process for resolving URI references that might be in relative form, along with guidelines and security considerations for the use of URIs on the Internet. The URI syntax defines a grammar that is a superset of all valid URIs, allowing an implementation to parse the common components of a URI reference without knowing the scheme-specific requirements of every possible identifier. This specification does not define a generative grammar for URIs; that task is performed by the individual specifications of each URI scheme. [STANDARDS-TRACK] This document provides guidelines and recommendations for the definition of Uniform Resource Identifier (URI) schemes. It also updates the process and IANA registry for URI schemes. It obsoletes both RFC 2717 and RFC 2718. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements. Many protocols make use of identifiers consisting of constants and other well-known values. Even after a protocol has been defined and deployment has begun, new values may need to be assigned (e.g., for a new option type in DHCP, or a new encryption or authentication transform for IPsec). To ensure that such quantities have consistent values and interpretations across all implementations, their assignment must be administered by a central authority. For IETF protocols, that role is provided by the Internet Assigned Numbers Authority (IANA).In order for IANA to manage a given namespace prudently, it needs guidelines describing the conditions under which new values can be assigned or when modifications to existing values can be made. If IANA is expected to play a role in the management of a namespace, IANA must be given clear and concise instructions describing that role. This document discusses issues that should be considered in formulating a policy for assigning values to a namespace and provides guidelines for authors on the specific text that must be included in documents that place demands on IANA.This document obsoletes RFC 2434. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements. This document specifies Version 1.2 of the Transport Layer Security (TLS) protocol. The TLS protocol provides communications security over the Internet. The protocol allows client/server applications to communicate in a way that is designed to prevent eavesdropping, tampering, or message forgery. [STANDARDS-TRACK] This memo defines a path prefix for "well-known locations", "/.well-known/", in selected Uniform Resource Identifier (URI) schemes. [STANDARDS-TRACK] The WebSocket Protocol enables two-way communication between a client running untrusted code in a controlled environment to a remote host that has opted-in to communications from that code. The security model used for this is the origin-based security model commonly used by web browsers. The protocol consists of an opening handshake followed by basic message framing, layered over TCP. The goal of this technology is to provide a mechanism for browser-based applications that need two-way communication with servers that does not rely on opening multiple HTTP connections (e.g., using XMLHttpRequest or <iframe>s and long polling). [STANDARDS-TRACK] The Constrained Application Protocol (CoAP) is a specialized web transfer protocol for use with constrained nodes and constrained (e.g., low-power, lossy) networks. The nodes often have 8-bit microcontrollers with small amounts of ROM and RAM, while constrained networks such as IPv6 over Low-Power Wireless Personal Area Networks (6LoWPANs) often have high packet error rates and a typical throughput of 10s of kbit/s. The protocol is designed for machine- to-machine (M2M) applications such as smart energy and building automation.CoAP provides a request/response interaction model between application endpoints, supports built-in discovery of services and resources, and includes key concepts of the Web such as URIs and Internet media types. CoAP is designed to easily interface with HTTP for integration with the Web while meeting specialized requirements such as multicast support, very low overhead, and simplicity for constrained environments. This document describes a Transport Layer Security (TLS) extension for application-layer protocol negotiation within the TLS handshake. For instances in which multiple application protocols are supported on the same TCP or UDP port, this extension allows the application layer to negotiate which protocol will be used within the TLS connection. Transport Layer Security (TLS) and Datagram Transport Layer Security (DTLS) are widely used to protect data exchanged over application protocols such as HTTP, SMTP, IMAP, POP, SIP, and XMPP. Over the last few years, several serious attacks on TLS have emerged, including attacks on its most commonly used cipher suites and their modes of operation. This document provides recommendations for improving the security of deployed services that use TLS and DTLS. The recommendations are applicable to the majority of use cases. This document updates the guidelines and recommendations, as well as the IANA registration processes, for the definition of Uniform Resource Identifier (URI) schemes. It obsoletes RFC 4395. The Constrained Application Protocol (CoAP) is a RESTful application protocol for constrained nodes and networks. The state of a resource on a CoAP server can change over time. This document specifies a simple protocol extension for CoAP that enables CoAP clients to "observe" resources, i.e., to retrieve a representation of a resource and keep this representation updated by the server over a period of time. The protocol follows a best-effort approach for sending new representations to clients and provides eventual consistency between the state observed by each client and the actual resource state at the server. A common design pattern in Internet of Things (IoT) deployments is the use of a constrained device that collects data via sensors or controls actuators for use in home automation, industrial control systems, smart cities and other IoT deployments. This document defines a Transport Layer Security (TLS) and Datagram TLS (DTLS) 1.2 profile that offers communications security for this data exchange thereby preventing eavesdropping, tampering, and message forgery. The lack of communication security is a common vulnerability in Internet of Things products that can easily be solved by using these well-researched and widely deployed Internet security protocols. CoAP is a RESTful transfer protocol for constrained nodes and networks. Basic CoAP messages work well for the small payloads we expect from temperature sensors, light switches, and similar building-automation devices. Occasionally, however, applications will need to transfer larger payloads -- for instance, for firmware updates. With HTTP, TCP does the grunt work of slicing large payloads up into multiple packets and ensuring that they all arrive and are handled in the right order. CoAP is based on datagram transports such as UDP or DTLS, which limits the maximum size of resource representations that can be transferred without too much fragmentation. Although UDP supports larger payloads through IP fragmentation, it is limited to 64 KiB and, more importantly, doesn't really work well for constrained applications and networks. Instead of relying on IP fragmentation, this specification extends basic CoAP with a pair of "Block" options, for transferring multiple blocks of information from a resource representation in multiple request-response pairs. In many important cases, the Block options enable a server to be truly stateless: the server can handle each block transfer separately, with no need for a connection setup or other server-side memory of previous block transfers. In summary, the Block options provide a minimal way to transfer larger representations in a block-wise fashion. Short Message Service (SMS) of mobile cellular networks is frequently used in Machine-To-Machine (M2M) communications, such as for telematic devices. The service offers small packet sizes and high delays just as other typical low-power and lossy networks (LLNs), i.e. 6LoWPANs. The design of the Constrained Application Protocol (CoAP) [RFC7252], that took the limitations of LLNs into account, is thus also applicable to other transports. The adaptation of CoAP to SMS transport mechanisms is described in this document. Internet technical specifications often need to define a formal syntax. Over the years, a modified version of Backus-Naur Form (BNF), called Augmented BNF (ABNF), has been popular among many Internet specifications. The current specification documents ABNF. It balances compactness and simplicity with reasonable representational power. The differences between standard BNF and ABNF involve naming rules, repetition, alternatives, order-independence, and value ranges. This specification also supplies additional rule definitions and encoding for a core lexical analyzer of the type common to several Internet specifications. [STANDARDS-TRACK] This document defines the concept of an "origin", which is often used as the scope of authority or privilege by user agents. Typically, user agents isolate content retrieved from different origins to prevent malicious web site operators from interfering with the operation of benign web sites. In addition to outlining the principles that underlie the concept of origin, this document details how to determine the origin of a URI and how to serialize an origin into a string. It also defines an HTTP header field, named "Origin", that indicates which origins are associated with an HTTP request. [STANDARDS-TRACK] This document defines the procedures that the Internet Assigned Numbers Authority (IANA) uses when handling assignment and other requests related to the Service Name and Transport Protocol Port Number registry. It also discusses the rationale and principles behind these procedures and how they facilitate the long-term sustainability of the registry.This document updates IANA's procedures by obsoleting the previous UDP and TCP port assignment procedures defined in Sections 8 and 9.1 of the IANA Allocation Guidelines, and it updates the IANA service name and port assignment procedures for UDP-Lite, the Datagram Congestion Control Protocol (DCCP), and the Stream Control Transmission Protocol (SCTP). It also updates the DNS SRV specification to clarify what a service name is and how it is registered. This memo documents an Internet Best Current Practice. This document specifies version 1.2 of the Datagram Transport Layer Security (DTLS) protocol. The DTLS protocol provides communications privacy for datagram protocols. The protocol allows client/server applications to communicate in a way that is designed to prevent eavesdropping, tampering, or message forgery. The DTLS protocol is based on the Transport Layer Security (TLS) protocol and provides equivalent security guarantees. Datagram semantics of the underlying transport are preserved by the DTLS protocol. This document updates DTLS 1.0 to work with TLS version 1.2. [STANDARDS-TRACK] The Hypertext Transfer Protocol (HTTP) is a stateless application-level protocol for distributed, collaborative, hypertext information systems. This document provides an overview of HTTP architecture and its associated terminology, defines the "http" and "https" Uniform Resource Identifier (URI) schemes, defines the HTTP/1.1 message syntax and parsing requirements, and describes related security concerns for implementations.

CoAP is defined in with a version number of 1. At this time, there is no known reason to support version numbers different from 1. In contrast to the message layer for UDP and DTLS, the CoAP over TCP message format does not include a version number. If version negotiation needs to be addressed in the future, then Capability and Settings have been specifically designed to enable such a potential feature.

This section gives examples for the first two configurations discussed in . An example of the process followed by a CoAP client to retrieve the representation of a resource identified by a “coap+ws” URI might be as follows. below illustrates the WebSocket and CoAP messages exchanged in detail. The CoAP client obtains the URI <coap+ws://example.org/sensors/temperature?u=Cel>, for example, from a resource representation that it retrieved previously. It establishes a WebSocket connection to the endpoint URI composed of the authority “example.org” and the well-known path “/.well-known/coap”, <ws://example.org/.well-known/coap>. It sends a single-frame, masked, binary message containing a CoAP request. The request indicates the target resource with the Uri-Path (“sensors”, “temperature”) and Uri-Query (“u=Cel”) options. It waits for the server to return a response. The CoAP client uses the connection for further requests, or the connection is closed.

| GET /.well-known/coap HTTP/1.1 | | Host: example.org | | Upgrade: websocket | | Connection: Upgrade | | Sec-WebSocket-Key: dGhlIHNhbXBsZSBub25jZQ== | | Sec-WebSocket-Protocol: coap | | Sec-WebSocket-Version: 13 | | |<=========+ HTTP/1.1 101 Switching Protocols | | Upgrade: websocket | | Connection: Upgrade | | Sec-WebSocket-Accept: s3pPLMBiTxaQ9kYGzzhZRbK+xOo= | | Sec-WebSocket-Protocol: coap | | | | +--------->| Binary frame (opcode=%x2, FIN=1, MASK=1) | | +-------------------------+ | | | GET | | | | Token: 0x53 | | | | Uri-Path: "sensors" | | | | Uri-Path: "temperature" | | | | Uri-Query: "u=Cel" | | | +-------------------------+ | | |<---------+ Binary frame (opcode=%x2, FIN=1, MASK=0) | | +-------------------------+ | | | 2.05 Content | | | | Token: 0x53 | | | | Payload: "22.3 Cel" | | | +-------------------------+ : : : : | | +--------->| Close frame (opcode=%x8, FIN=1, MASK=1) | | |<---------+ Close frame (opcode=%x8, FIN=1, MASK=0) | | ]]>

shows how a CoAP client uses a CoAP forward proxy with a WebSocket endpoint to retrieve the representation of the resource “coap://[2001:DB8::1]/”. The use of the forward proxy and the address of the WebSocket endpoint are determined by the client from local configuration rules. The request URI is specified in the Proxy-Uri Option. Since the request URI uses the “coap” URI scheme, the proxy fulfills the request by issuing a Confirmable GET request over UDP to the CoAP server and returning the response over the WebSocket connection to the client.

| | Binary frame (opcode=%x2, FIN=1, MASK=1) | | | +------------------------------------+ | | | | GET | | | | | Token: 0x7d | | | | | Proxy-Uri: "coap://[2001:DB8::1]/" | | | | +------------------------------------+ | | | | +--------->| CoAP message (Ver=1, T=Con, MID=0x8f54) | | | +------------------------------------+ | | | | GET | | | | | Token: 0x0a15 | | | | +------------------------------------+ | | | | |<---------+ CoAP message (Ver=1, T=Ack, MID=0x8f54) | | | +------------------------------------+ | | | | 2.05 Content | | | | | Token: 0x0a15 | | | | | Payload: "ready" | | | | +------------------------------------+ | | | |<---------+ | Binary frame (opcode=%x2, FIN=1, MASK=0) | | | +------------------------------------+ | | | | 2.05 Content | | | | | Token: 0x7d | | | | | Payload: "ready" | | | | +------------------------------------+ | | | ]]>

The RFC Editor is requested to remove this section at publication.

Merged draft-savolainen-core-coap-websockets-07 Merged draft-bormann-core-block-bert-01 Merged draft-bormann-core-coap-sig-02

We would like to thank Stephen Berard, Geoffrey Cristallo, Olivier Delaby, Christian Groves, Nadir Javed, Michael Koster, Matthias Kovatsch, Achim Kraus, David Navarro, Szymon Sasin, Zach Shelby, Andrew Summers, Julien Vermillard, and Gengyu Wei for their feedback.