path: root/lib/kernel/doc/src/gen_sctp.xml

<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE erlref SYSTEM "erlref.dtd">

<erlref>
  <header>
    <copyright>
      <year>2007</year><year>2018</year>
      <holder>Ericsson AB. All Rights Reserved.</holder>
    </copyright>
    <legalnotice>
      Licensed under the Apache License, Version 2.0 (the "License");
      you may not use this file except in compliance with the License.
      You may obtain a copy of the License at
 
          http://www.apache.org/licenses/LICENSE-2.0

      Unless required by applicable law or agreed to in writing, software
      distributed under the License is distributed on an "AS IS" BASIS,
      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
      See the License for the specific language governing permissions and
      limitations under the License.

    </legalnotice>

    <title>gen_sctp</title>
    <prepared>[email protected]</prepared>
    <responsible></responsible>
    <docno>1</docno>
    <approved></approved>
    <checked></checked>
    <date>2007-03-21</date>
    <rev>A</rev>
    <file>gen_sctp.xml</file>
  </header>
  <module since="">gen_sctp</module>
  <modulesummary>Functions for communicating with sockets using the SCTP
    protocol.</modulesummary>
  <description>
    <p>This module provides functions for communicating with
      sockets using the SCTP protocol. The implementation assumes that
      the OS kernel supports SCTP
      <url href="http://www.rfc-archive.org/getrfc.php?rfc=2960">(RFC 2960)</url>
      through the user-level
      <url href="http://tools.ietf.org/html/draft-ietf-tsvwg-sctpsocket-13">Sockets API Extensions</url>.</p>
    <p>During development, this implementation was tested on:</p>
    <list type="bulleted">
      <item>Linux Fedora Core 5.0 (kernel 2.6.15-2054 or later is needed)</item>
      <item>Solaris 10, 11</item>
    </list>
    <p>During OTP adaptation it was tested on:</p>
    <list type="bulleted">
      <item>SUSE Linux Enterprise Server 10 (x86_64) kernel 2.6.16.27-0.6-smp,
        with lksctp-tools-1.0.6</item>
      <item>Briefly on Solaris 10</item>
      <item>SUSE Linux Enterprise Server 10 Service Pack 1 (x86_64)
        kernel 2.6.16.54-0.2.3-smp with lksctp-tools-1.0.7</item>
      <item>FreeBSD 8.2</item>
    </list>
    <p>This module was written for one-to-many style sockets
      (type <c>seqpacket</c>). With the addition of
      <seealso marker="#peeloff/2"><c>peeloff/2</c></seealso>,
      one-to-one style sockets (type <c>stream</c>) were introduced.</p>
    <p>Record definitions for this module can be found using:</p>
    <pre>
-include_lib("kernel/include/inet_sctp.hrl").</pre>
    <p>These record definitions use the "new" spelling 'adaptation',
      not the deprecated 'adaption', regardless of which
      spelling the underlying C API uses.</p>
  </description>

  <datatypes>
    <datatype>
      <name>assoc_id()</name>
      <desc>
        <p>An opaque term returned in, for example, <c>#sctp_paddr_change{}</c>,
          which identifies an association for an SCTP socket. The term
          is opaque except for the special value <c>0</c>, which has a
          meaning such as "the whole endpoint" or "all future associations".</p>
      </desc>
    </datatype>
    <datatype>
      <name name="option"/>
      <desc>
        <p>One of the
          <seealso marker="#options">SCTP Socket Options</seealso>.</p>
      </desc>
    </datatype>
    <datatype>
      <name name="option_name"/>
    </datatype>
    <datatype>
      <name>sctp_socket()</name>
      <desc>
        <p>Socket identifier returned from
          <seealso marker="#open/0"><c>open/*</c></seealso>.</p>
        <marker id="exports"></marker>
      </desc>
    </datatype>
  </datatypes>

  <funcs>
    <func>
      <name name="abort" arity="2" since=""/>
      <fsummary>Abnormally terminate the association specified by
        <c>Assoc</c>, without flushing of unsent data.</fsummary>
      <desc>
        <p>Abnormally terminates the association specified by
          <c><anno>Assoc</anno></c>, without
          flushing of unsent data. The socket itself remains open. Other
          associations opened on this socket are still valid, and the socket
          can be used in new associations.</p>
      </desc>
    </func>

    <func>
      <name name="close" arity="1" since=""/>
      <fsummary>Close the socket and all associations on it.</fsummary>
      <desc>
        <p>Closes the socket and all associations on it. The unsent
          data is flushed as in <seealso marker="#eof/2"><c>eof/2</c></seealso>.
          The <c>close/1</c> call
          is blocking or otherwise depending of the value of
          the <seealso marker="inet#option-linger"><c>linger</c></seealso>
          socket <seealso marker="#options">option</seealso>.
          If <c>close</c> does not linger or linger time-out expires,
          the call returns and the data is flushed in the background.</p>
      </desc>
    </func>

    <func>
      <name name="connect" arity="4" since=""/>
      <fsummary>Same as <c>connect(Socket, Addr, Port, Opts, infinity)</c>.</fsummary>
      <desc>
        <p>Same as <c>connect(<anno>Socket</anno>, <anno>Addr</anno>,
          <anno>Port</anno>, <anno>Opts</anno>, infinity)</c>.</p>
      </desc>
    </func>

    <func>
      <name name="connect" arity="5" since=""/>
      <fsummary>Establish a new association for socket <c>Socket</c>, with a
        peer (SCTP server socket).</fsummary>
      <desc>
        <p>Establishes a new association for socket <c><anno>Socket</anno></c>,
          with the peer (SCTP server socket) specified by
          <c><anno>Addr</anno></c> and <c><anno>Port</anno></c>.
          <c><anno>Timeout</anno></c>, is expressed in milliseconds.
          A socket can be associated with multiple peers.</p>
        <warning><p>Using a value of <c><anno>Timeout</anno></c> less than
          the maximum time taken by the OS to establish an association (around
          4.5 minutes if the default values from
          <url href="https://tools.ietf.org/html/rfc4960">RFC 4960</url>
          are used), can result
          in inconsistent or incorrect return values. This is especially
          relevant for associations sharing the same <c><anno>Socket</anno></c>
          (that is, source address and port), as the controlling process
          blocks until <c>connect/*</c> returns.
          <seealso marker="#connect_init/4"><c>connect_init/*</c></seealso>
          provides an alternative without this limitation.</p>
        </warning>
        <p><marker id="record-sctp_assoc_change"></marker>
          The result of <c>connect/*</c> is an <c>#sctp_assoc_change{}</c>
          event that contains, in particular, the new
          <seealso marker="#type-assoc_id">Association ID</seealso>:</p>
        <pre>
#sctp_assoc_change{
      state             = atom(),
      error             = atom(),
      outbound_streams  = integer(),
      inbound_streams   = integer(),
      assoc_id          = assoc_id()
}</pre>
        <p>The number of outbound and inbound streams can be set by
          giving an <c>sctp_initmsg</c> option to <c>connect</c> as in:</p>
        <pre>
connect(Socket, Ip, Port>,
      [{sctp_initmsg,#sctp_initmsg{num_ostreams=OutStreams,
                                   max_instreams=MaxInStreams}}])</pre>
        <p>All options <c><anno>Opt</anno></c> are set on the socket before the
          association is attempted. If an option record has undefined
          field values, the options record is first read from the socket
          for those values. In effect, <c><anno>Opt</anno></c> option records
          only define field values to change before connecting.</p>
        <p>The returned <c>outbound_streams</c> and <c>inbound_streams</c>
          are the stream numbers on the socket. These can be different
          from the requested values (<c>OutStreams</c> and <c>MaxInStreams</c>,
          respectively) if the peer requires lower values.</p>
        <p><c>state</c> can have the following values:</p>
        <taglist>
          <tag><c>comm_up</c></tag>
          <item><p>Association is successfully established. This
            indicates a successful completion of <c>connect</c>.</p></item>
          <tag><c>cant_assoc</c></tag>
          <item><p>The association cannot be established
            (<c>connect/*</c> failure).</p></item>
        </taglist>
        <p>Other states do not normally occur in the output from
          <c>connect/*</c>. Rather, they can occur in
          <c>#sctp_assoc_change{}</c> events received instead of data in
          <seealso marker="#recv/1"><c>recv/*</c></seealso> calls.
          All of them indicate losing the association because of various error
          conditions, and are listed here for the sake of completeness:</p>
        <taglist>
          <tag><c>comm_lost</c></tag>
          <item></item>
          <tag><c>restart</c></tag>
          <item></item>
          <tag><c>shutdown_comp</c></tag>
          <item></item>
        </taglist>
        <p>Field <c>error</c> can provide more detailed diagnostics.</p>
      </desc>
    </func>

    <func>
      <name name="connect_init" arity="4" since="OTP R13B04"/>
      <fsummary>Same as <c>connect_init(Socket, Addr, Port, Opts, infinity)</c>..</fsummary>
      <desc>
        <p>Same as <c>connect_init(<anno>Socket</anno>, <anno>Addr</anno>,
          <anno>Port</anno>, <anno>Opts</anno>, infinity)</c>.</p>
      </desc>
    </func>

    <func>
      <name name="connect_init" arity="5" since="OTP R13B04"/>
      <fsummary>Initiate a new association for socket <c>Socket</c>, with a
        peer (SCTP server socket).</fsummary>
      <desc>
        <p>Initiates a new association for socket <c><anno>Socket</anno></c>,
          with the peer (SCTP server socket) specified by
          <c><anno>Addr</anno></c> and <c><anno>Port</anno></c>.</p>
        <p>The fundamental difference between this API
          and <c>connect/*</c> is that the return value is that of the
          underlying OS <c>connect(2)</c> system call. If <c>ok</c> is returned,
          the result of the association establishment is received
          by the calling process as an
          <seealso marker="#record-sctp_assoc_change"><c>#sctp_assoc_change{}</c></seealso>
          event. The calling process must be prepared to receive this, or
          poll for it using
          <seealso marker="#recv/1"><c>recv/*</c></seealso>,
          depending on the value of the active option.</p>
        <p>The parameters are as described in
          <seealso marker="#connect/5"><c>connect/*</c></seealso>,
           except the <c><anno>Timeout</anno></c> value.</p>
        <p>The timer associated with <c><anno>Timeout</anno></c> only supervises
          IP resolution of <c><anno>Addr</anno></c>.</p>
      </desc>
    </func>

    <func>
      <name name="controlling_process" arity="2" since=""/>
      <fsummary>Assign a new controlling process pid to the socket.</fsummary>
      <desc>
        <p>Assigns a new controlling process <c><anno>Pid</anno></c> to
          <c><anno>Socket</anno></c>. Same implementation as
          <seealso marker="gen_udp#controlling_process/2"><c>gen_udp:controlling_process/2</c></seealso>.
        </p>
      </desc>
    </func>

    <func>
      <name name="eof" arity="2" since=""/>
      <fsummary>Gracefully terminate the association specified by <c>Assoc</c>,
        with flushing of all unsent data.</fsummary>
      <desc>
        <p>Gracefully terminates the association specified by
          <c><anno>Assoc</anno></c>, with
          flushing of all unsent data. The socket itself remains open. Other
          associations opened on this socket are still valid. The socket can
          be used in new associations.</p>
      </desc>
    </func>

    <func>
      <name name="error_string" arity="1" since=""/>
      <fsummary>Translate an SCTP error number into a string.</fsummary>
      <desc>
        <p>Translates an SCTP error number from, for example,
          <c>#sctp_remote_error{}</c> or <c>#sctp_send_failed{}</c> into
          an explanatory string, or one of the atoms <c>ok</c> for no
          error or <c>undefined</c> for an unrecognized error.</p>
      </desc>
    </func>

    <func>
      <name name="listen" arity="2" clause_i="1" since=""/>
      <name name="listen" arity="2" clause_i="2" since="OTP R15B"/>
      <fsummary>Set up a socket to listen.</fsummary>
      <desc>
        <p>Sets up a socket to listen on the IP address and port number
          it is bound to.</p>
        <p>For type <c>seqpacket</c>, sockets (the default)
          <c><anno>IsServer</anno></c> must be <c>true</c> or <c>false</c>.
          In contrast to TCP, there is no listening queue length in SCTP.
          If <c><anno>IsServer</anno></c> is <c>true</c>, the socket accepts
          new associations, that is, it becomes an SCTP server socket.</p>
	<p>For type <c>stream</c>, sockets <anno>Backlog</anno> define
          the backlog queue length just like in TCP.</p>
      </desc>
    </func>

    <func>
      <name name="open" arity="0" since=""/>
      <name name="open" arity="1" clause_i="1" since=""/>
      <name name="open" arity="1" clause_i="2" since=""/>
      <name name="open" arity="2" since=""/>
      <fsummary>Create an SCTP socket and binds it to local addresses.</fsummary>
      <desc>
        <p>Creates an SCTP socket and binds it to the local addresses
          specified by all <c>{ip,<anno>IP</anno>}</c> (or synonymously
          <c>{ifaddr,<anno>IP</anno>}</c>)
          options (this feature is called SCTP multi-homing). The default
          <c><anno>IP</anno></c> and <c><anno>Port</anno></c> are <c>any</c>
          and <c>0</c>, meaning bind to all local addresses on any
          free port.</p>
        <p>Other options:</p>
        <taglist>
          <tag><c>inet6</c></tag>
          <item>
            <p>Sets up the socket for IPv6.</p>
          </item>
          <tag><c>inet</c></tag>
          <item>
            <p>Sets up the socket for IPv4. This is the default.</p>
          </item>
        </taglist>
        <p>A default set of socket <seealso marker="#options">options</seealso>
          is used. In particular, the socket is opened in
          <seealso marker="#option-binary">binary</seealso> and
          <seealso marker="#option-active">passive</seealso> mode,
	  with <anno>SockType</anno> <c>seqpacket</c>, and with reasonably large
          <seealso marker="inet#option-sndbuf">kernel</seealso> and driver
          <seealso marker="inet#option-buffer">buffers</seealso>.</p>
	<p>
	  If the socket is in
          <seealso marker="#option-active">passive</seealso>
	  mode data can be received through the
          <seealso marker="#recv/1"><c>recv/1,2</c></seealso>
	  calls.
	</p>
	<p>
	  If the socket is in
          <seealso marker="#option-active">active</seealso>
	  mode data received data is delivered to the controlling process
	  as messages:
	</p>
        <code type="none">
{sctp, <anno>Socket</anno>, FromIP, FromPort, {AncData, Data}}
	</code>
	<p>
	  See
	  <seealso marker="#recv/1"><c>recv/1,2</c></seealso>
	  for a description of the message fields.
	</p>
	<note>
	  <p>
	    This message format unfortunately differs slightly from the
	    <seealso marker="gen_udp#open/1"><c>gen_udp</c></seealso>
	    message format with ancillary data,
	    and from the
	    <seealso marker="#recv/1"><c>recv/1,2</c></seealso>
	    return tuple format.
	  </p>
	</note>
      </desc>
    </func>

    <func>
      <name name="peeloff" arity="2" since="OTP R15B"/>
      <fsummary>Peel off a type <c>stream</c> socket from a type
        <c>seqpacket</c> one.</fsummary>
      <desc>
        <p>Branches off an existing association <c><anno>Assoc</anno></c>
          in a socket <c><anno>Socket</anno></c> of type <c>seqpacket</c>
          (one-to-many style) into
          a new socket <c><anno>NewSocket</anno></c> of type <c>stream</c>
          (one-to-one style).</p>
        <p>The existing association argument <c><anno>Assoc</anno></c>
          can be either a
          <seealso marker="#record-sctp_assoc_change"><c>#sctp_assoc_change{}</c></seealso>
          record as returned from, for example,
          <seealso marker="#recv-2"><c>recv/*</c></seealso>,
          <seealso marker="#connect-5"><c>connect/*</c></seealso>, or
	  from a listening socket in active mode. It can also be just
	  the field <c>assoc_id</c> integer from such a record.</p>
      </desc>
    </func>

    <func>
      <name name="recv" arity="1" since=""/>
      <name name="recv" arity="2" since=""/>
      <fsummary>Receive a message from a socket.</fsummary>
      <desc>
        <p>Receives the <c><anno>Data</anno></c> message from any association
          of the socket.
          If the receive times out, <c>{error,timeout}</c> is returned.
          The default time-out is <c>infinity</c>. <c><anno>FromIP</anno></c>
          and <c><anno>FromPort</anno></c> indicate the address of the
          sender.</p>
        <p><c><anno>AncData</anno></c> is a list of ancillary data items that
          can be received along with the main <c><anno>Data</anno></c>.
          This list can be empty, or contain a single
          <seealso marker="#record-sctp_sndrcvinfo"><c>#sctp_sndrcvinfo{}</c></seealso>
          record if receiving of such ancillary data is enabled (see option
          <seealso marker="#option-sctp_events"><c>sctp_events</c></seealso>).
          It is enabled by default, as such ancillary data
          provides an easy way of determining the association and stream
          over which the message is received.
          (An alternative way is to get the association ID from
          <c><anno>FromIP</anno></c> and <c><anno>FromPort</anno></c> using
          socket option
          <seealso marker="#option-sctp_get_peer_addr_info"><c>sctp_get_peer_addr_info</c></seealso>,
          but this does still not produce the stream number).</p>
	<p>
	  <c><anno>AncData</anno></c> may also contain
	  <seealso marker="inet#type-ancillary_data">
	    ancillary data
	  </seealso>
	  from the socket
	  <seealso marker="#type-option">options</seealso>
	  <seealso marker="inet#option-recvtos"><c>recvtos</c></seealso>,
	  <seealso marker="inet#option-recvtclass"><c>recvtclass</c></seealso>
	  or
	  <seealso marker="inet#option-recvttl"><c>recvttl</c></seealso>,
	  if that is supported by the platform for the socket.
	</p>
        <p>The <c><anno>Data</anno></c> received can be a <c>binary()</c>
          or a <c>list()</c> of bytes (integers in the range 0 through 255)
          depending on the socket mode, or an SCTP event.</p>
        <marker id="sctp_events"></marker>
        <p>Possible SCTP events:</p>
        <list type="bulleted">
          <item>
            <seealso marker="#record-sctp_sndrcvinfo"><c>#sctp_sndrcvinfo{}</c></seealso>
          </item>
          <item>
            <seealso marker="#record-sctp_assoc_change"><c>#sctp_assoc_change{}</c></seealso>
          </item>
          <item>
            <pre>
#sctp_paddr_change{
      addr      = {ip_address(),port()},
      state     = atom(),
      error     = integer(),
      assoc_id  = assoc_id()
}</pre>
            <p>Indicates change of the status of the IP address of the peer
              specified by
              <c>addr</c> within association <c>assoc_id</c>. Possible
              values of <c>state</c> (mostly self-explanatory) include:</p>
            <taglist>
              <tag><c>addr_unreachable</c></tag>
              <item></item>
              <tag><c>addr_available</c></tag>
              <item></item>
              <tag><c>addr_removed</c></tag>
              <item></item>
              <tag><c>addr_added</c></tag>
              <item></item>
              <tag><c>addr_made_prim</c></tag>
              <item></item>
              <tag><c>addr_confirmed</c></tag>
              <item></item>
            </taglist>
            <p>In case of an error (for example, <c>addr_unreachable</c>),
              field <c>error</c> provides more diagnostics. In such cases,
              event <c>#sctp_paddr_change{}</c> is automatically
              converted into an <c>error</c> term returned by
              <seealso marker="#recv/1"><c>recv</c></seealso>.
              The <c>error</c> field value can be converted into a string using
              <seealso marker="#error_string/1"><c>error_string/1</c></seealso>.
            </p>
          </item>
          <item>
          <pre>
#sctp_send_failed{
      flags     = true | false,
      error     = integer(),
      info      = #sctp_sndrcvinfo{},
      assoc_id  = assoc_id()
      data      = binary()
}</pre>
            <p>The sender can receive this event if a send operation fails.</p>
            <taglist>
              <tag><c>flags</c></tag>
              <item><p>A Boolean specifying if the data has been transmitted
                over the wire.</p></item>
              <tag><c>error</c></tag>
              <item><p>Provides extended diagnostics, use
                <seealso marker="#error_string/1"><c>error_string/1</c>.</seealso></p>
              </item>
              <tag><c>info</c></tag>
              <item><p>The original
                <seealso marker="#record-sctp_sndrcvinfo"><c>#sctp_sndrcvinfo{}</c></seealso>
                record used in the failed
                <seealso marker="#send/3"><c>send/*</c>.</seealso></p>
              </item>
              <tag><c>data</c></tag>
              <item><p>The whole original data chunk attempted to be sent.</p>
              </item>
            </taglist>
            <p>In the current implementation of the Erlang/SCTP binding,
              this event is internally converted into an <c>error</c> term
              returned by
              <seealso marker="#recv/1"><c>recv/*</c></seealso>.</p>
          </item>
          <item>
          <pre>
#sctp_adaptation_event{
      adaptation_ind = integer(),
      assoc_id       = assoc_id()
}</pre>
            <p>Delivered when a peer sends an adaptation layer indication
              parameter (configured through option
              <seealso marker="#option-sctp_adaptation_layer"><c>sctp_adaptation_layer</c></seealso>).
              Notice that with the current implementation of
              the Erlang/SCTP binding, this event is disabled by default.</p>
          </item>
          <item>
          <pre>
#sctp_pdapi_event{
      indication = sctp_partial_delivery_aborted,
      assoc_id   = assoc_id()
}</pre>
            <p>A partial delivery failure. In the current implementation of
              the Erlang/SCTP binding, this event is internally converted
              into an <c>error</c> term returned by
              <seealso marker="#recv/1"><c>recv/*</c></seealso>.</p>
          </item>
        </list>
      </desc>
    </func>

    <func>
      <name name="send" arity="3" since=""/>
      <fsummary>Send a message using an <c>#sctp_sndrcvinfo{}</c>record.</fsummary>
      <desc>
        <p>Sends the <c><anno>Data</anno></c> message with all sending
          parameters from a
          <seealso marker="#record-sctp_sndrcvinfo"><c>#sctp_sndrcvinfo{}</c></seealso>
          record. This way, the user can specify the PPID (passed to the remote
          end) and context (passed to the local SCTP layer), which can be used,
          for example, for error identification.
          However, such a fine level of user control is rarely required.
          The function <c>send/4</c> is sufficient for most applications.</p>
      </desc>
    </func>

    <func>
      <name name="send" arity="4" since=""/>
      <fsummary>Send a message over an existing association and specified
        stream.</fsummary>
      <desc>
        <p>Sends a <c><anno>Data</anno></c> message over an existing association
          and specified stream.</p>
      </desc>
    </func>
  </funcs>

  <section>
    <marker id="options"></marker>
    <title>SCTP Socket Options</title>
    <p>The set of admissible SCTP socket options is by construction
      orthogonal to the sets of TCP, UDP, and generic <c>inet</c> options.
      Only options listed here are allowed
      for SCTP sockets. Options can be set on the socket using
      <seealso marker="#open/1"><c>open/1,2</c></seealso> or
      <seealso marker="inet#setopts/2"><c>inet:setopts/2</c></seealso>,
      retrieved using
      <seealso marker="inet#getopts/2"><c>inet:getopts/2</c></seealso>.
      Options can be changed when calling
      <seealso marker="#connect/4"><c>connect/4,5</c></seealso>.</p>
    <marker id="option-binary"></marker>
    <marker id="option-list"></marker>
    <taglist>
      <tag><c>{mode, list|binary}</c> or just <c>list</c> or <c>binary</c></tag>
      <item>
        <p>Determines the type of data returned from
          <seealso marker="#recv/1"><c>recv/1,2</c></seealso>.</p>
        <marker id="option-active"></marker>
      </item>
      <tag><c>{active, true|false|once|N}</c></tag>
      <item>
        <list type="bulleted">
          <item>
            <p>If <c>false</c> (passive mode, the default),
              the caller must do an explicit
              <seealso marker="#recv/1"><c>recv</c></seealso> call
              to retrieve the available data from the socket.</p>
          </item>
	  <item>
	    <p>
	      If <c>true|once|N</c> (active modes)
	      received data or events are sent to the owning process.
	      See <seealso marker="#open/0"><c>open/0..2</c></seealso>
	      for the message format.
	    </p>
	  </item>
          <item>
            <p>
	      If <c>true</c> (full active mode) there is no flow control.
	    </p>
	    <note>
              <p>
		Note that this can cause the message queue to overflow
		causing for example the virtual machine
		to run out of memory and crash.
	      </p>
	    </note>
          </item>
          <item>
            <p>If <c>once</c>, only one message is automatically placed
              in the message queue, and after that the mode is automatically
              reset to passive. This provides flow control and
              the possibility for the receiver to listen for its incoming
              SCTP data interleaved with other inter-process messages.</p>
          </item>
          <item>
            <p>If <c>active</c> is specified as an integer <c>N</c> in the
              range -32768 to 32767 (inclusive), that number is added to
              the socket's counting of data messages to be
              delivered to the controlling process. If the result of the
              addition is negative, the count is set to <c>0</c>. Once the
              count reaches <c>0</c>, either through the delivery of messages
              or by being explicitly set with
              <seealso marker="inet#setopts/2"><c>inet:setopts/2</c></seealso>,
              the socket mode is automatically reset to passive (<c>{active,
              false}</c>). When a socket in this active mode transitions to
              passive mode, the message <c>{sctp_passive, Socket}</c> is sent
              to the controlling process to notify it that if it wants to
              receive more data messages from the socket, it must call
              <seealso marker="inet#setopts/2"><c>inet:setopts/2</c></seealso>
              to set the socket back into an active mode.</p>
          </item>
        </list>
      </item>
      <tag><c>{tos, integer()}</c></tag>
      <item>
        <p>Sets the Type-Of-Service field on the IP datagrams that are sent,
          to the specified value. This effectively determines a prioritization
          policy for the outbound packets. The acceptable values
          are system-dependent.</p>
      </item>
      <tag><c>{priority, integer()}</c></tag>
      <item>
        <p>A protocol-independent equivalent of <c>tos</c> above. Setting
          priority implies setting <c>tos</c> as well.</p>
      </item>
      <tag><c>{dontroute, true|false}</c></tag>
      <item>
        <p>Defaults to <c>false</c>. If <c>true</c>, the kernel does not
          send packets through any gateway, only sends them to directly
          connected hosts.</p>
      </item>
      <tag><c>{reuseaddr, true|false}</c></tag>
      <item>
        <p>Defaults to <c>false</c>. If true, the local binding address
          <c>{IP,Port}</c> of the socket can be reused immediately.
          No waiting in state <c>CLOSE_WAIT</c> is performed (can  be
          required for high-throughput servers).</p>
      </item>
      <tag><c>{sndbuf, integer()}</c></tag>
      <item>
        <p>The size, in bytes, of the OS kernel send buffer for this socket.
          Sending errors would occur for datagrams larger than
          <c>val(sndbuf)</c>. Setting this option also adjusts
          the size of the driver buffer (see <c>buffer</c> above).</p>
      </item>
      <tag><c>{recbuf, integer()}</c></tag>
      <item>
        <p>The size, in bytes, of the OS kernel receive buffer for this socket.
          Sending errors would occur for datagrams larger than
          <c>val(recbuf)</c>. Setting this option also adjusts
          the size of the driver buffer (see <c>buffer</c> above).</p>
      </item>
      <tag><c>{sctp_module, module()}</c></tag>
      <item>
        <p>Overrides which callback module is used. Defaults to
          <c>inet_sctp</c> for IPv4 and <c>inet6_sctp</c> for IPv6.</p>
      </item>
      <tag><c>{sctp_rtoinfo, #sctp_rtoinfo{}}</c></tag>
      <item>
      <pre>
#sctp_rtoinfo{
      assoc_id = assoc_id(),
      initial  = integer(),
      max      = integer(),
      min      = integer()
}</pre>
        <p>Determines retransmission time-out parameters, in milliseconds,
          for the association(s) specified by <c>assoc_id</c>.</p>
        <p><c>assoc_id = 0</c> (default) indicates the whole endpoint. See
          <url href="http://www.rfc-archive.org/getrfc.php?rfc=2960">RFC
          2960</url> and
          <url href="http://tools.ietf.org/html/draft-ietf-tsvwg-sctpsocket-13">Sockets
          API Extensions for SCTP</url>
          for the exact semantics of the field values.</p>
      </item>
      <tag><c>{sctp_associnfo, #sctp_assocparams{}}</c></tag>
      <item>
      <pre>
#sctp_assocparams{
      assoc_id                 = assoc_id(),
      asocmaxrxt               = integer(),
      number_peer_destinations = integer(),
      peer_rwnd                = integer(),
      local_rwnd               = integer(),
      cookie_life              = integer()
}</pre>
        <p>Determines association parameters for the association(s) specified by
          <c>assoc_id</c>.</p>
        <p><c>assoc_id = 0</c> (default) indicates the whole endpoint. See
          <url href="http://tools.ietf.org/html/draft-ietf-tsvwg-sctpsocket-13">Sockets API Extensions for SCTP</url>
          for the discussion of their semantics. Rarely used.</p>
      </item>
      <tag><c>{sctp_initmsg, #sctp_initmsg{}}</c></tag>
      <item>
      <pre>
#sctp_initmsg{
     num_ostreams   = integer(),
     max_instreams  = integer(),
     max_attempts   = integer(),
     max_init_timeo = integer()
}</pre>
        <p>Determines the default parameters that this socket tries
          to negotiate with its peer while establishing an association with it.
          Is to be set after
          <seealso marker="#open/1"><c>open/*</c></seealso>
          but before the first
          <seealso marker="#connect/4"><c>connect/*</c></seealso>.
          <c>#sctp_initmsg{}</c> can also be used
          as ancillary data with the first call of
          <seealso marker="#send/3"><c>send/*</c></seealso> to
          a new peer (when a new association is created).</p>
        <taglist>
          <tag><c>num_ostreams</c></tag>
          <item>Number of outbound streams</item>
          <tag><c>max_instreams</c></tag>
          <item>Maximum number of inbound streams</item>
          <tag><c>max_attempts</c></tag>
          <item>Maximum retransmissions while establishing an association</item>
          <tag><c>max_init_timeo</c></tag>
          <item>Time-out, in milliseconds, for establishing an association</item>
        </taglist>
      </item>
      <tag><c>{sctp_autoclose, integer() >= 0}</c></tag>
      <item>
        <p>Determines the time, in seconds, after which an idle association is
          automatically closed. <c>0</c> means that the association is
          never automatically closed.</p>
      </item>
      <tag><c>{sctp_nodelay, true|false}</c></tag>
      <item>
        <p>Turns on|off the Nagle algorithm for merging small packets
          into larger ones. This improves throughput at the expense
          of latency.</p>
      </item>
      <tag><c>{sctp_disable_fragments, true|false}</c></tag>
      <item>
        <p>If <c>true</c>, induces an error on an attempt to send
          a message larger than the current PMTU size
          (which would require fragmentation/reassembling).
          Notice that message fragmentation does not affect
          the logical atomicity of its delivery; this option
          is provided for performance reasons only.</p>
      </item>
      <tag><c>{sctp_i_want_mapped_v4_addr, true|false}</c></tag>
      <item>
        <p>Turns on|off automatic mapping of IPv4 addresses into IPv6 ones
          (if the socket address family is <c>AF_INET6</c>).</p>
      </item>
      <tag><c>{sctp_maxseg, integer()}</c></tag>
      <item>
        <p>Determines the maximum chunk size if message fragmentation is used.
          If <c>0</c>, the chunk size is limited by the Path MTU only.</p>
      </item>
      <tag><c>{sctp_primary_addr, #sctp_prim{}}</c></tag>
      <item>
      <pre>
#sctp_prim{
      assoc_id = assoc_id(),
      addr     = {IP, Port}
}
 IP = ip_address()
 Port = port_number()</pre>
        <p>For the association specified by <c>assoc_id</c>,
          <c>{IP,Port}</c> must be one of the peer addresses.
          This option determines that the specified address is treated by
          the local SCTP stack as the primary address of the peer.</p>
      </item>
      <tag><c>{sctp_set_peer_primary_addr, #sctp_setpeerprim{}}</c></tag>
      <item>
      <pre>
#sctp_setpeerprim{
      assoc_id = assoc_id(),
      addr     = {IP, Port}
}
 IP = ip_address()
 Port = port_number()</pre>
        <p>When set, informs the peer to use <c>{IP, Port}</c>
          as the primary address of the local endpoint for the association
          specified by <c>assoc_id</c>.</p>
        <marker id="option-sctp_adaptation_layer"></marker>
      </item>
      <tag><c>{sctp_adaptation_layer, #sctp_setadaptation{}}</c></tag>
      <item>
        <marker id="record-sctp_setadaptation"></marker>
      <pre>
#sctp_setadaptation{
      adaptation_ind = integer()
}</pre>
        <p>When set, requests that the local endpoint uses the value specified
          by <c>adaptation_ind</c> as the Adaptation Indication parameter for
          establishing new associations. For details, see
          <url href="http://www.rfc-archive.org/getrfc.php?rfc=2960">RFC 2960</url>
          and
          <url href="http://tools.ietf.org/html/draft-ietf-tsvwg-sctpsocket-13">Sockets
          API Extenstions for SCTP</url>.</p>
      </item>
      <tag><c>{sctp_peer_addr_params, #sctp_paddrparams{}}</c></tag>
      <item>
        <pre>
#sctp_paddrparams{
      assoc_id   = assoc_id(),
      address    = {IP, Port},
      hbinterval = integer(),
      pathmaxrxt = integer(),
      pathmtu    = integer(),
      sackdelay  = integer(),
      flags      = list()
}
IP = ip_address()
Port = port_number()</pre>
        <p>Determines various per-address parameters for
          the association specified by <c>assoc_id</c> and the peer address
          <c>address</c> (the SCTP protocol supports multi-homing, so
          more than one address can correspond to a specified association).</p>
        <taglist>
          <tag><c>hbinterval</c></tag>
          <item><p>Heartbeat interval, in milliseconds</p></item>
          <tag><c>pathmaxrxt</c></tag>
          <item><p>Maximum number of retransmissions before this address is
            considered unreachable (and an alternative address is selected)</p>
          </item>
          <tag><c>pathmtu</c></tag>
          <item><p>Fixed Path MTU, if automatic discovery is disabled (see
            <c>flags</c> below)</p></item>
          <tag><c>sackdelay</c></tag>
          <item><p>Delay, in milliseconds, for SAC messages (if the delay is
            enabled, see <c>flags</c> below)</p></item>
          <tag><c>flags</c></tag>
          <item><p>The following flags are available:</p>
            <taglist>
              <tag><c>hb_enable</c></tag>
              <item>Enables heartbeat</item>
              <tag><c>hb_disable</c></tag>
              <item>Disables heartbeat</item>
              <tag><c>hb_demand</c></tag>
              <item>Initiates heartbeat immediately</item>
              <tag><c>pmtud_enable</c></tag>
              <item>Enables automatic Path MTU discovery</item>
              <tag><c>pmtud_disable</c></tag>
              <item>Disables automatic Path MTU discovery</item>
              <tag><c>sackdelay_enable</c></tag>
              <item>Enables SAC delay</item>
              <tag><c>sackdelay_disable</c></tag>
              <item>Disables SAC delay</item>
            </taglist></item>
        </taglist>
      </item>
      <tag><c>{sctp_default_send_param, #sctp_sndrcvinfo{}}</c></tag>
      <item>
        <marker id="record-sctp_sndrcvinfo"></marker>
        <pre>
#sctp_sndrcvinfo{
      stream     = integer(),
      ssn        = integer(),
      flags      = list(),
      ppid       = integer(),
      context    = integer(),
      timetolive = integer(),
      tsn        = integer(),
      cumtsn     = integer(),
      assoc_id   = assoc_id()
}</pre>
        <p><c>#sctp_sndrcvinfo{}</c> is used both in this socket option, and as
          ancillary data while sending or receiving SCTP messages. When
          set as an option, it provides default values for subsequent
          <seealso marker="#send/3"><c>send</c></seealso>
          calls on the association specified by
          <c>assoc_id</c>.</p>
        <p><c>assoc_id = 0</c> (default) indicates
          the whole endpoint.</p>
        <p>The following fields typically must be specified by the sender:</p>
        <taglist>
          <tag><c>sinfo_stream</c></tag>
          <item><p>Stream number (0-base) within the association
            to send the messages through;</p></item>
          <tag><c>sinfo_flags</c></tag>
          <item><p>The following flags are recognised:</p>
            <taglist>
              <tag><c>unordered</c></tag>
              <item>The message is to be sent unordered</item>
              <tag><c>addr_over</c></tag>
              <item>The address specified in
                <seealso marker="#send/3"><c>send</c></seealso>
                overwrites the primary peer address</item>
              <tag><c>abort</c></tag>
              <item>Aborts the current association without flushing any unsent
                data</item>
              <tag><c>eof</c></tag>
              <item>Gracefully shuts down the current association, with
                flushing of unsent data</item>
            </taglist>
              <p>Other fields are rarely used. For complete information, see
                <url href="http://www.rfc-archive.org/getrfc.php?rfc=2960">RFC 2960</url>
                and
                <url href="http://tools.ietf.org/html/draft-ietf-tsvwg-sctpsocket-13">Sockets
                API Extensions for SCTP</url>.</p></item>
        </taglist>
        <marker id="option-sctp_events"></marker>
      </item>
      <tag><c>{sctp_events, #sctp_event_subscribe{}}</c></tag>
      <item>
        <marker id="record-sctp_event_subscribe"></marker>
        <pre>
#sctp_event_subscribe{
        data_io_event          = true | false,
        association_event      = true | false,
        address_event          = true | false,
        send_failure_event     = true | false,
        peer_error_event       = true | false,
        shutdown_event         = true | false,
        partial_delivery_event = true | false,
        adaptation_layer_event = true | false
}</pre>
        <p>This option determines which
          <seealso marker="#sctp_events">SCTP Events</seealso> are to be
          received (through
          <seealso marker="#recv/1"><c>recv/*</c></seealso>)
          along with the data. The only exception is <c>data_io_event</c>,
          which enables or disables receiving of
          <seealso marker="#record-sctp_sndrcvinfo"><c>#sctp_sndrcvinfo{}</c></seealso>
          ancillary data, not events.
          By default, all flags except <c>adaptation_layer_event</c> are
          enabled, although <c>sctp_data_io_event</c> and
          <c>association_event</c> are used by the driver itself and not
          exported to the user level.</p>
      </item>
      <tag><c>{sctp_delayed_ack_time, #sctp_assoc_value{}}</c></tag>
      <item>
        <pre>
#sctp_assoc_value{
      assoc_id    = assoc_id(),
      assoc_value = integer()
}</pre>
        <p>Rarely used. Determines the ACK time
          (specified by <c>assoc_value</c>, in milliseconds) for
          the specified association or the whole endpoint
          if <c>assoc_value = 0</c> (default).</p>
      </item>
      <tag><c>{sctp_status, #sctp_status{}}</c></tag>
      <item>
        <pre>
#sctp_status{
      assoc_id            = assoc_id(),
      state               = atom(),
      rwnd                = integer(),
      unackdata           = integer(),
      penddata            = integer(),
      instrms             = integer(),
      outstrms            = integer(),
      fragmentation_point = integer(),
      primary             = #sctp_paddrinfo{}
}</pre>
        <p>This option is read-only. It determines the status of
          the SCTP association specified by <c>assoc_id</c>.
          The following are the
          possible values of <c>state</c> (the state designations are mostly
          self-explanatory):</p>
        <taglist>
          <tag><c>sctp_state_empty</c></tag>
          <item>Default. Means that no other state is active.</item>
          <tag><c>sctp_state_closed</c></tag>
          <item></item>
          <tag><c>sctp_state_cookie_wait</c></tag>
          <item></item>
          <tag><c>sctp_state_cookie_echoed</c></tag>
          <item></item>
          <tag><c>sctp_state_established</c></tag>
          <item></item>
          <tag><c>sctp_state_shutdown_pending</c></tag>
          <item></item>
          <tag><c>sctp_state_shutdown_sent</c></tag>
          <item></item>
          <tag><c>sctp_state_shutdown_received</c></tag>
          <item></item>
          <tag><c>sctp_state_shutdown_ack_sent</c></tag>
          <item></item>
        </taglist>
        <p>Semantics of the other fields:</p>
        <taglist>
          <tag><c>sstat_rwnd</c></tag>
          <item>Current receiver window size of the association</item>
          <tag><c>sstat_unackdata</c></tag>
          <item>Number of unacked data chunks</item>
          <tag><c>sstat_penddata</c></tag>
          <item>Number of data chunks pending receipt</item>
          <tag><c>sstat_instrms</c></tag>
          <item>Number of inbound streams</item>
          <tag><c>sstat_outstrms</c></tag>
          <item>Number of outbound streams</item>
          <tag><c>sstat_fragmentation_point</c></tag>
          <item>Message size at which SCTP fragmentation occurs</item>
          <tag><c>sstat_primary</c></tag>
          <item>Information on the current primary peer address (see below for
            the format of <c>#sctp_paddrinfo{}</c>)</item>
        </taglist>
        <marker id="option-sctp_get_peer_addr_info"></marker>
      </item>
      <tag><c>{sctp_get_peer_addr_info, #sctp_paddrinfo{}}</c></tag>
      <item>
        <marker id="record-sctp_paddrinfo"></marker>
        <pre>
#sctp_paddrinfo{
      assoc_id  = assoc_id(),
      address   = {IP, Port},
      state     = inactive | active | unconfirmed,
      cwnd      = integer(),
      srtt      = integer(),
      rto       = integer(),
      mtu       = integer()
}
IP = ip_address()
Port = port_number()</pre>
        <p>This option is read-only. It determines the parameters specific to
          the peer address specified by <c>address</c> within the association
          specified by <c>assoc_id</c>. Field <c>address</c> fmust be set by the
          caller; all other fields are filled in on return.
          If <c>assoc_id = 0</c> (default), the <c>address</c>
          is automatically translated into the corresponding
          association ID. This option is rarely used.
          For the semantics of all fields, see
          <url href="http://www.rfc-archive.org/getrfc.php?rfc=2960">RFC 2960</url>
          and
          <url href="http://tools.ietf.org/html/draft-ietf-tsvwg-sctpsocket-13">Sockets
          API Extensions for SCTP</url>.</p>
      </item>
    </taglist>
  </section>

  <section>
    <marker id="examples"></marker>
    <title>SCTP Examples</title>
      <p>Example of an Erlang SCTP server that receives SCTP messages and
         prints them on the standard output:</p>
      <pre>
-module(sctp_server).

-export([server/0,server/1,server/2]).
-include_lib("kernel/include/inet.hrl").
-include_lib("kernel/include/inet_sctp.hrl").

server() -&gt;
    server(any, 2006).

server([Host,Port]) when is_list(Host), is_list(Port) -&gt;
    {ok, #hostent{h_addr_list = [IP|_]}} = inet:gethostbyname(Host),
    io:format("~w -&gt; ~w~n", [Host, IP]),
    server([IP, list_to_integer(Port)]).

server(IP, Port) when is_tuple(IP) orelse IP == any orelse IP == loopback,
                      is_integer(Port) -&gt;
    {ok,S} = gen_sctp:open(Port, [{recbuf,65536}, {ip,IP}]),
    io:format("Listening on ~w:~w. ~w~n", [IP,Port,S]),
    ok     = gen_sctp:listen(S, true),
    server_loop(S).

server_loop(S) -&gt;
    case gen_sctp:recv(S) of
    {error, Error} -&gt;
        io:format("SCTP RECV ERROR: ~p~n", [Error]);
    Data -&gt;
        io:format("Received: ~p~n", [Data])
    end,
    server_loop(S).</pre>
      <p>Example of an Erlang SCTP client interacting with the above server.
        Notice that in this example the client creates an association with
        the server with 5 outbound streams. Therefore, sending of
        <c>"Test 0"</c> over stream 0 succeeds, but sending of <c>"Test 5"</c>
        over stream 5 fails. The client then <c>abort</c>s the association,
        which results in that the corresponding event is received on
        the server side.</p>
      <pre>
-module(sctp_client).

-export([client/0, client/1, client/2]).
-include_lib("kernel/include/inet.hrl").
-include_lib("kernel/include/inet_sctp.hrl").

client() -&gt;
    client([localhost]).

client([Host]) -&gt;
    client(Host, 2006);
 
client([Host, Port]) when is_list(Host), is_list(Port) -&gt;
    client(Host,list_to_integer(Port)),
    init:stop().

client(Host, Port) when is_integer(Port) -&gt;
    {ok,S}     = gen_sctp:open(),
    {ok,Assoc} = gen_sctp:connect
        (S, Host, Port, [{sctp_initmsg,#sctp_initmsg{num_ostreams=5}}]),
    io:format("Connection Successful, Assoc=~p~n", [Assoc]),

    io:write(gen_sctp:send(S, Assoc, 0, &lt;&lt;"Test 0"&gt;&gt;)),
    io:nl(),
    timer:sleep(10000),
    io:write(gen_sctp:send(S, Assoc, 5, &lt;&lt;"Test 5"&gt;&gt;)),
    io:nl(),
    timer:sleep(10000),
    io:write(gen_sctp:abort(S, Assoc)),
    io:nl(),

    timer:sleep(1000),
    gen_sctp:close(S).</pre>
      <p>A simple Erlang SCTP client that uses the <c>connect_init</c> API:</p>
      <pre>
-module(ex3).

-export([client/4]).
-include_lib("kernel/include/inet.hrl").
-include_lib("kernel/include/inet_sctp.hrl").

client(Peer1, Port1, Peer2, Port2)
  when is_tuple(Peer1), is_integer(Port1), is_tuple(Peer2), is_integer(Port2) -&gt;
    {ok,S}     = gen_sctp:open(),
    SctpInitMsgOpt = {sctp_initmsg,#sctp_initmsg{num_ostreams=5}},
    ActiveOpt = {active, true},
    Opts = [SctpInitMsgOpt, ActiveOpt],
    ok = gen_sctp:connect(S, Peer1, Port1, Opts),
    ok = gen_sctp:connect(S, Peer2, Port2, Opts),
    io:format("Connections initiated~n", []),
    client_loop(S, Peer1, Port1, undefined, Peer2, Port2, undefined).

client_loop(S, Peer1, Port1, AssocId1, Peer2, Port2, AssocId2) -&gt;
    receive
        {sctp, S, Peer1, Port1, {_Anc, SAC}}
          when is_record(SAC, sctp_assoc_change), AssocId1 == undefined -&gt;
            io:format("Association 1 connect result: ~p. AssocId: ~p~n",
                      [SAC#sctp_assoc_change.state,
                       SAC#sctp_assoc_change.assoc_id]),
            client_loop(S, Peer1, Port1, SAC#sctp_assoc_change.assoc_id,
                        Peer2, Port2, AssocId2);

        {sctp, S, Peer2, Port2, {_Anc, SAC}}
          when is_record(SAC, sctp_assoc_change), AssocId2 == undefined -&gt;
            io:format("Association 2 connect result: ~p. AssocId: ~p~n",
                      [SAC#sctp_assoc_change.state, SAC#sctp_assoc_change.assoc_id]),
            client_loop(S, Peer1, Port1, AssocId1, Peer2, Port2,
                       SAC#sctp_assoc_change.assoc_id);

        {sctp, S, Peer1, Port1, Data} -&gt;
            io:format("Association 1: received ~p~n", [Data]),
            client_loop(S, Peer1, Port1, AssocId1,
                        Peer2, Port2, AssocId2);

        {sctp, S, Peer2, Port2, Data} -&gt;
            io:format("Association 2: received ~p~n", [Data]),
            client_loop(S, Peer1, Port1, AssocId1,
                        Peer2, Port2, AssocId2);

        Other -&gt;
            io:format("Other ~p~n", [Other]),
            client_loop(S, Peer1, Port1, AssocId1,
                        Peer2, Port2, AssocId2)

    after 5000 -&gt;
            ok
    end.</pre>
  </section>

  <section>
    <marker id="seealso"></marker>
    <title>See Also</title>
    <p><seealso marker="gen_tcp"><c>gen_tcp(3)</c></seealso>,
      <seealso marker="gen_udp"><c>gen_udp(3)</c></seealso>,
      <seealso marker="inet"><c>inet(3)</c></seealso>,
      <url href="http://www.rfc-archive.org/getrfc.php?rfc=2960">RFC 2960</url>
      (Stream Control Transmission Protocol),
      <url href="http://tools.ietf.org/html/draft-ietf-tsvwg-sctpsocket-13">Sockets
      API Extensions for SCTP</url></p>
  </section>
</erlref>