<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE chapter SYSTEM "chapter.dtd">
<chapter>
<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.
The Initial Developer of the Original Code is Ericsson AB.
</legalnotice>
<title>Distribution Protocol</title>
<prepared></prepared>
<docno></docno>
<date>2007-09-21</date>
<rev>PA1</rev>
<file>erl_dist_protocol.xml</file>
</header>
<p>This description is far from complete. It will be updated if the
protocol is updated. However, the protocols, both from Erlang
nodes to the Erlang Port Mapper Daemon (EPMD) and between Erlang nodes
are stable since many years.</p>
<p>The distribution protocol can be divided into four parts:</p>
<list type="bulleted">
<item>
<p>Low-level socket connection (1)</p>
</item>
<item>
<p>Handshake, interchange node name, and authenticate (2)</p>
</item>
<item>
<p>Authentication (done by <seealso marker="kernel:net_kernel">
<c>net_kernel(3)</c></seealso>) (3)</p>
</item>
<item>
<p>Connected (4)</p>
</item>
</list>
<p>A node fetches the port number of another node through the EPMD (at the
other host) to initiate a connection request.</p>
<p>For each host, where a distributed Erlang node is running, also an EPMD
is to be running. The EPMD can be started explicitly or automatically
as a result of the Erlang node startup.</p>
<p>By default the EPMD listens on port 4369.</p>
<p>(3) and (4) above are performed at the same level but the <c>net_kernel</c>
disconnects the other node if it communicates using an invalid cookie (after
1 second).</p>
<p>The integers in all multibyte fields are in big-endian order.</p>
<warning>
<p>
The Erlang Distribution protocol is not by itself secure and does not
aim to be so. In order to get secure distribution the distributed nodes
should be configured to use distribution over tls.
See the <seealso marker="ssl:ssl_distribution">
Using SSL for Erlang Distribution</seealso> User's Guide
for details on how to setup a secure distributed node.
</p>
</warning>
<section>
<title>EPMD Protocol</title>
<p>The requests served by the EPMD are summarized in the following
figure.</p>
<image file="erl_ext_fig.gif">
<icaption>Summary of EPMD Requests</icaption>
</image>
<p>Each request <c>*_REQ</c> is preceded by a 2 byte length field.
Thus, the overall request format is as follows:</p>
<table align="left">
<row>
<cell align="center">2</cell>
<cell align="center">n</cell>
</row>
<row>
<cell align="center"><c>Length</c></cell>
<cell align="center"><c>Request</c></cell>
</row>
<tcaption>Request Format</tcaption>
</table>
<section>
<title>Register a Node in EPMD</title>
<p>When a distributed node is started it registers itself in the EPMD.
The message <c>ALIVE2_REQ</c> described below is sent from the node to
the EPMD. The response from the EPMD is <c>ALIVE2_X_RESP</c> (or
<c>ALIVE2_RESP</c>).</p>
<table align="left">
<row>
<cell align="center">1</cell>
<cell align="center">2</cell>
<cell align="center">1</cell>
<cell align="center">1</cell>
<cell align="center">2</cell>
<cell align="center">2</cell>
<cell align="center">2</cell>
<cell align="center">Nlen</cell>
<cell align="center">2</cell>
<cell align="center">Elen</cell>
</row>
<row>
<cell align="center"><c>120</c></cell>
<cell align="center"><c>PortNo</c></cell>
<cell align="center"><c>NodeType</c></cell>
<cell align="center"><c>Protocol</c></cell>
<cell align="center"><c>HighestVersion</c></cell>
<cell align="center"><c>LowestVersion</c></cell>
<cell align="center"><c>Nlen</c></cell>
<cell align="center"><c>NodeName</c></cell>
<cell align="center"><c>Elen</c></cell>
<cell align="center"><c>Extra</c></cell>
</row>
<tcaption>ALIVE2_REQ (120)</tcaption>
</table>
<taglist>
<tag><c>PortNo</c></tag>
<item>
<p>The port number on which the node accept connection requests.</p>
</item>
<tag><c>NodeType</c></tag>
<item>
<p>77 = normal Erlang node, 72 = hidden node (C-node), ...</p>
</item>
<tag><c>Protocol</c></tag>
<item>
<p>0 = TCP/IPv4, ...</p>
</item>
<tag><c>HighestVersion</c></tag>
<item>
<p>The highest distribution version that this node can handle.
The value in OTP 22 and later is 6.</p>
</item>
<tag><c>LowestVersion</c></tag>
<item>
<p>The lowest distribution version that this node can handle.
The value in OTP 22 and later is 5.</p>
</item>
<tag><c>Nlen</c></tag>
<item>
<p>The length (in bytes) of field <c>NodeName</c>.</p>
</item>
<tag><c>NodeName</c></tag>
<item>
<p>The node name as an UTF-8 encoded string of <c>Nlen</c> bytes.</p>
</item>
<tag><c>Elen</c></tag>
<item>
<p>The length of field <c>Extra</c>.</p>
</item>
<tag><c>Extra</c></tag>
<item>
<p>Extra field of <c>Elen</c> bytes.</p>
</item>
</taglist>
<p>The connection created to the EPMD must be kept as long as the
node is a distributed node. When the connection is closed,
the node is automatically unregistered from the EPMD.</p>
<p>The response message is either <c>ALIVE2_X_RESP</c> or
<c>ALIVE2_RESP</c> depending on distribution version. If both the node
and EPMD support distribution version 6 then response is
<c>ALIVE2_X_RESP</c> otherwise it is the older <c>ALIVE2_RESP</c>:</p>
<table align="left">
<row>
<cell align="center">1</cell>
<cell align="center">1</cell>
<cell align="center">4</cell>
</row>
<row>
<cell align="center"><c>118</c></cell>
<cell align="center"><c>Result</c></cell>
<cell align="center"><c>Creation</c></cell>
</row>
<tcaption>ALIVE2_X_RESP (118) with 32 bit creation</tcaption>
</table>
<table align="left">
<row>
<cell align="center">1</cell>
<cell align="center">1</cell>
<cell align="center">2</cell>
</row>
<row>
<cell align="center"><c>121</c></cell>
<cell align="center"><c>Result</c></cell>
<cell align="center"><c>Creation</c></cell>
</row>
<tcaption>ALIVE2_RESP (121) with 16-bit creation</tcaption>
</table>
<p>Result = 0 -> ok, result > 0 -> error.</p>
</section>
<section>
<title>Unregister a Node from EPMD</title>
<p>A node unregisters itself from the EPMD by closing the TCP
connection to EPMD established when the node was registered.</p>
</section>
<section>
<title>Get the Distribution Port of Another Node</title>
<p>When one node wants to connect to another node it starts with
a <c>PORT_PLEASE2_REQ</c> request to the EPMD on the host where the
node resides to get the distribution port that the node listens to.</p>
<table align="left">
<row>
<cell align="center">1</cell>
<cell align="center">N</cell>
</row>
<row>
<cell align="center"><c>122</c></cell>
<cell align="center"><c>NodeName</c></cell>
</row>
<tcaption>PORT_PLEASE2_REQ (122)</tcaption>
</table>
<p>where N = <c>Length</c> - 1.</p>
<table align="left">
<row>
<cell align="center">1</cell>
<cell align="center">1</cell>
</row>
<row>
<cell align="center"><c>119</c></cell>
<cell align="center"><c>Result</c></cell>
</row>
<tcaption>PORT2_RESP (119) Response Indicating Error, Result > 0
</tcaption>
</table>
<p>or</p>
<table align="left">
<row>
<cell align="center">1</cell>
<cell align="center">1</cell>
<cell align="center">2</cell>
<cell align="center">1</cell>
<cell align="center">1</cell>
<cell align="center">2</cell>
<cell align="center">2</cell>
<cell align="center">2</cell>
<cell align="center">Nlen</cell>
<cell align="center">2</cell>
<cell align="center">Elen</cell>
</row>
<row>
<cell align="center"><c>119</c></cell>
<cell align="center"><c>Result</c></cell>
<cell align="center"><c>PortNo</c></cell>
<cell align="center"><c>NodeType</c></cell>
<cell align="center"><c>Protocol</c></cell>
<cell align="center"><c>HighestVersion</c></cell>
<cell align="center"><c>LowestVersion</c></cell>
<cell align="center"><c>Nlen</c></cell>
<cell align="center"><c>NodeName</c></cell>
<cell align="center"><c>Elen</c></cell>
<cell align="center">><c>Extra</c></cell>
</row>
<tcaption>PORT2_RESP, Result = 0</tcaption>
</table>
<p>If <c>Result</c> > 0, the packet only consists of
<c>[119, Result]</c>.</p>
<p>The EPMD closes the socket when it has sent the information.</p>
</section>
<section>
<title>Get All Registered Names from EPMD</title>
<p>This request is used through the Erlang function
<seealso marker="kernel:net_adm#names/1,2">
<c>net_adm:names/1,2</c></seealso>. A TCP connection is opened
to the EPMD and this request is sent.</p>
<table align="left">
<row>
<cell align="center">1</cell>
</row>
<row>
<cell align="center"><c>110</c></cell>
</row>
<tcaption>NAMES_REQ (110)</tcaption>
</table>
<p>The response for a <c>NAMES_REQ</c> is as follows:</p>
<table align="left">
<row>
<cell align="center">4</cell>
<cell align="center"> </cell>
</row>
<row>
<cell align="center"><c>EPMDPortNo</c></cell>
<cell align="center"><c>NodeInfo*</c></cell>
</row>
<tcaption>NAMES_RESP</tcaption>
</table>
<p><c>NodeInfo</c> is a string written for each active node.
When all <c>NodeInfo</c> has been written the connection is
closed by the EPMD.</p>
<p><c>NodeInfo</c> is, as expressed in Erlang:</p>
<code>
io:format("name ~ts at port ~p~n", [NodeName, Port]).</code>
</section>
<section>
<title>Dump All Data from EPMD</title>
<p>This request is not really used, it is to be regarded as a debug
feature.</p>
<table align="left">
<row>
<cell align="center">1</cell>
</row>
<row>
<cell align="center"><c>100</c></cell>
</row>
<tcaption>DUMP_REQ</tcaption>
</table>
<p>The response for a <c>DUMP_REQ</c> is as follows:</p>
<table align="left">
<row>
<cell align="center">4</cell>
<cell align="center"> </cell>
</row>
<row>
<cell align="center"><c>EPMDPortNo</c></cell>
<cell align="center"><c>NodeInfo*</c></cell>
</row>
<tcaption>DUMP_RESP</tcaption>
</table>
<p><c>NodeInfo</c> is a string written for each node kept in the EPMD.
When all <c>NodeInfo</c> has been written the connection is
closed by the EPMD.</p>
<p><c>NodeInfo</c> is, as expressed in Erlang:</p>
<code>
io:format("active name ~ts at port ~p, fd = ~p~n",
[NodeName, Port, Fd]).</code>
<p>or</p>
<code>
io:format("old/unused name ~ts at port ~p, fd = ~p ~n",
[NodeName, Port, Fd]).</code>
</section>
<section>
<title>Kill EPMD</title>
<p>This request kills the running EPMD. It is almost never used.</p>
<table align="left">
<row>
<cell align="center">1</cell>
</row>
<row>
<cell align="center"><c>107</c></cell>
</row>
<tcaption>KILL_REQ</tcaption>
</table>
<p>The response for a <c>KILL_REQ</c> is as follows:</p>
<table align="left">
<row>
<cell align="center">2</cell>
</row>
<row>
<cell align="center"><c>OKString</c></cell>
</row>
<tcaption>KILL_RESP</tcaption>
</table>
<p>where <c>OKString</c> is "OK".</p>
</section>
<section>
<title>STOP_REQ (Not Used)</title>
<table align="left">
<row>
<cell align="center">1</cell>
<cell align="center">n</cell>
</row>
<row>
<cell align="center"><c>115</c></cell>
<cell align="center"><c>NodeName</c></cell>
</row>
<tcaption>STOP_REQ</tcaption>
</table>
<p>where n = <c>Length</c> - 1.</p>
<p>The current implementation of Erlang does not care if the connection
to the EPMD is broken.</p>
<p>The response for a <c>STOP_REQ</c> is as follows:</p>
<table align="left">
<row>
<cell align="center">7</cell>
</row>
<row>
<cell align="center"><c>OKString</c></cell>
</row>
<tcaption>STOP_RESP</tcaption>
</table>
<p>where <c>OKString</c> is "STOPPED".</p>
<p>A negative response can look as follows:</p>
<table align="left">
<row>
<cell align="center">7</cell>
</row>
<row>
<cell align="center"><c>NOKString</c></cell>
</row>
<tcaption>STOP_NOTOK_RESP</tcaption>
</table>
<p>where <c>NOKString</c> is "NOEXIST".</p>
</section>
<!--
<section>
<title>ALIVE_REQ (97)</title>
<table align="left">
<row>
<cell align="center">1</cell>
<cell align="center">2</cell>
<cell align="center">n</cell>
</row>
<row>
<cell align="center"><c>97</c></cell>
<cell align="center"><c>PortNo</c></cell>
<cell align="center"><c>NodeName</c></cell>
</row>
<tcaption></tcaption>
</table>
<p>where n = <c>Length</c> - 3.</p>
<p>The connection created to the EPMD must be kept until the node is
not a distributed node any longer.</p>
</section>
<section>
<title>ALIVE_OK_RESP (89)</title>
<table align="left">
<row>
<cell align="center">1</cell>
<cell align="center">2</cell>
</row>
<row>
<cell align="center"><c>89</c></cell>
<cell align="center"><c>Creation</c></cell>
</row>
<tcaption></tcaption>
</table>
</section>
<section>
<title>ALIVE_NOTOK_RESP ()</title>
<p>The EPMD closed the connection.</p>
</section>
<section>
<title>PORT_PLEASE_REQ (112)</title>
<table align="left">
<row>
<cell align="center">1</cell>
<cell align="center">n</cell>
</row>
<row>
<cell align="center"><c>112</c></cell>
<cell align="center"><c>NodeName</c></cell>
</row>
<tcaption></tcaption>
</table>
<p>where n = <c>Length</c> - 1.</p>
</section>
<section>
<title>PORT_OK_RESP ()</title>
<table align="left">
<row>
<cell align="center">2</cell>
</row>
<row>
<cell align="center"><c>PortNo</c></cell>
</row>
<tcaption></tcaption>
</table>
</section>
<section>
<title>PORT_NOTOK_RESP ()</title>
<p>The EPMD closed the connection.</p>
</section>
<section>
<title>PORT_NOTOK_RESP ()</title>
<p>The EPMD closed the connection.</p>
</section>
-->
</section>
<section>
<marker id="distribution_handshake"/>
<title>Distribution Handshake</title>
<p>This section describes the distribution handshake protocol introduced
in Erlang/OTP R6. The handshake has remained almost the same since then.</p>
<section>
<title>General</title>
<p>The TCP/IP distribution uses a handshake that expects a
connection-based protocol, that is, the protocol does not include any
authentication after the handshake procedure.</p>
<p>This is not entirely safe, as it is vulnerable against takeover
attacks, but it is a tradeoff between fair safety and performance.</p>
<p>The cookies are never sent in cleartext and the handshake procedure
expects the client (called <c>A</c>) to be the first one to prove that
it can generate a sufficient digest. The digest is generated with the
MD5 message digest algorithm and the challenges are expected to be
random numbers.</p>
</section>
<section>
<title>Definitions</title>
<p>A challenge is a 32-bit integer in big-endian order. Below the function
<c>gen_challenge()</c> returns a random 32-bit integer used as a
challenge.</p>
<p>A digest is a (16 bytes) MD5 hash of the challenge (as text)
concatenated with the cookie (as text). Below, the function
<c>gen_digest(Challenge, Cookie)</c> generates a digest as described
above.</p>
<p>An <c>out_cookie</c> is the cookie used in outgoing communication to a
certain node, so that <c>A</c>'s <c>out_cookie</c> for <c>B</c> is to
correspond with <c>B</c>'s <c>in_cookie</c> for <c>A</c> and conversely.
<c>A</c>'s <c>out_cookie</c> for <c>B</c> and <c>A</c>'s
<c>in_cookie</c> for <c>B</c> need <em>not</em> be the same. Below the
function <c>out_cookie(Node)</c> returns the current node's
<c>out_cookie</c> for <c>Node</c>.</p>
<p>An <c>in_cookie</c> is the cookie expected to be used by another node
when communicating with us, so that <c>A</c>'s <c>in_cookie</c> for
<c>B</c> corresponds with <c>B</c>'s <c>out_cookie</c> for <c>A</c>.
Below the function <c>in_cookie(Node)</c> returns the current node's
<c>in_cookie</c> for <c>Node</c>.</p>
<p>The cookies are text strings that can be viewed as passwords.</p>
<p>Every message in the handshake starts with a 16-bit big-endian integer,
which contains the message length (not counting the two initial bytes).
In Erlang this corresponds to option <c>{packet, 2}</c> in
<seealso marker="kernel:gen_tcp"><c>gen_tcp(3)</c></seealso>.
Notice that after the handshake, the distribution switches to 4 byte
packet headers.</p>
</section>
<section>
<title>The Handshake in Detail</title>
<p>Imagine two nodes, <c>A</c> that initiates the handshake and <c>B</c>
that accepts the connection.</p>
<taglist>
<tag>1) connect/accept</tag>
<item>
<p><c>A</c> connects to <c>B</c> through TCP/IP and <c>B</c> accepts
the connection.</p>
</item>
<tag>2) <c>send_name</c>/<c>receive_name</c></tag>
<item>
<p><c>A</c> sends an initial identification to <c>B</c>, which
receives the message. The message looks as follows (every "square"
is one byte and the packet header is removed):</p>
<pre>
+---+--------+--------+-----+-----+-----+-----+-----+-----+-...-+-----+
|'n'|Version0|Version1|Flag0|Flag1|Flag2|Flag3|Name0|Name1| ... |NameN|
+---+--------+--------+-----+-----+-----+-----+-----+-----+-... +-----+</pre>
<p>'n' is the message tag. 'Version0' and 'Version1' is the
distribution version selected by <c>A</c>, based on information
from the EPMD. (16-bit big-endian) 'Flag0' ... 'Flag3' are
capability flags, the capabilities are defined in
<c>$ERL_TOP/lib/kernel/include/dist.hrl</c>. (32-bit big-endian)
'Name0' ... 'NameN' is the full node name of <c>A</c>, as a string
of bytes (the packet length denotes how long it is).</p>
</item>
<tag>3) <c>recv_status</c>/<c>send_status</c></tag>
<item>
<p><c>B</c> sends a status message to <c>A</c>, which indicates if the
connection is allowed. The following status codes are defined:</p>
<taglist>
<tag><c>ok</c></tag>
<item>
<p>The handshake will continue.</p>
</item>
<tag><c>ok_simultaneous</c></tag>
<item>
<p>The handshake will continue, but <c>A</c> is informed that
<c>B</c> has another ongoing connection attempt that will be
shut down (simultaneous connect where <c>A</c>'s name is
greater than <c>B</c>'s name, compared literally).</p>
</item>
<tag><c>nok</c></tag>
<item>
<p>The handshake will not continue, as <c>B</c> already has an
ongoing handshake, which it itself has initiated (simultaneous
connect where <c>B</c>'s name is greater than <c>A</c>'s).</p>
</item>
<tag><c>not_allowed</c></tag>
<item>
<p>The connection is disallowed for some (unspecified) security
reason.</p>
</item>
<tag><c>alive</c></tag>
<item>
<p>A connection to the node is already active, which either means
that node <c>A</c> is confused or that the TCP connection
breakdown of a previous node with this name has not yet reached
node <c>B</c>. See step 3B below.</p>
</item>
</taglist>
<p>The format of the status message is as follows:</p>
<pre>
+---+-------+-------+-...-+-------+
|'s'|Status0|Status1| ... |StatusN|
+---+-------+-------+-...-+-------+</pre>
<p>'s' is the message tag. 'Status0' ... 'StatusN' is the status as a
string (not terminated).</p>
</item>
<tag>3B) <c>send_status</c>/<c>recv_status</c></tag>
<item>
<p>If status was <c>alive</c>, node <c>A</c> answers with another
status message containing either <c>true</c>, which means that the
connection is to continue (the old connection from this node is
broken), or <c>false</c>, which means that the connection is to be
closed (the connection attempt was a mistake.</p>
</item>
<tag>4) <c>recv_challenge</c>/<c>send_challenge</c></tag>
<item>
<p>If the status was <c>ok</c> or <c>ok_simultaneous</c>, the
handshake continues with <c>B</c> sending <c>A</c> another message,
the challenge. The challenge contains the same type of information
as the "name" message initially sent from <c>A</c> to <c>B</c>, plus
a 32-bit challenge:</p>
<pre>
+---+--------+--------+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-...-+-----+
|'n'|Version0|Version1|Flag0|Flag1|Flag2|Flag3|Chal0|Chal1|Chal2|Chal3|Name0|Name1| ... |NameN|
+---+--------+--------+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-... +-----+</pre>
<p>'Chal0' ... 'Chal3' is the challenge as a 32-bit big-endian integer
and the other fields are <c>B</c>'s version, flags, and full node
name.</p>
</item>
<tag>5) <c>send_challenge_reply</c>/<c>recv_challenge_reply</c></tag>
<item>
<p>Now <c>A</c> has generated a digest and its own challenge. Those
are sent together in a package to <c>B</c>:</p>
<pre>
+---+-----+-----+-----+-----+-----+-----+-----+-----+-...-+------+
|'r'|Chal0|Chal1|Chal2|Chal3|Dige0|Dige1|Dige2|Dige3| ... |Dige15|
+---+-----+-----+-----+-----+-----+-----+-----+-----+-...-+------+</pre>
<p>'r' is the tag. 'Chal0' ... 'Chal3' is <c>A</c>'s challenge for
<c>B</c> to handle. 'Dige0' ... 'Dige15' is the digest that <c>A</c>
constructed from the challenge <c>B</c> sent in the previous
step.</p>
</item>
<tag>6) <c>recv_challenge_ack</c>/<c>send_challenge_ack</c></tag>
<item>
<p><c>B</c> checks that the digest received from <c>A</c> is correct
and generates a digest from the challenge received from <c>A</c>.
The digest is then sent to <c>A</c>. The message is as follows:</p>
<pre>
+---+-----+-----+-----+-----+-...-+------+
|'a'|Dige0|Dige1|Dige2|Dige3| ... |Dige15|
+---+-----+-----+-----+-----+-...-+------+</pre>
<p>'a' is the tag. 'Dige0' ... 'Dige15' is the digest calculated by
<c>B</c> for <c>A</c>'s challenge.</p>
</item>
<tag>7) check</tag>
<item>
<p><c>A</c> checks the digest from <c>B</c> and the connection is
up.</p>
</item>
</taglist>
</section>
<section>
<title>Semigraphic View</title>
<pre>
A (initiator) B (acceptor)
TCP connect ------------------------------------>
TCP accept
send_name -------------------------------------->
recv_name
<---------------------------------------------- send_status
recv_status
(if status was 'alive'
send_status - - - - - - - - - - - - - - - - - ->
recv_status)
ChB = gen_challenge()
(ChB)
<---------------------------------------------- send_challenge
recv_challenge
ChA = gen_challenge(),
OCA = out_cookie(B),
DiA = gen_digest(ChB, OCA)
(ChA, DiA)
send_challenge_reply --------------------------->
recv_challenge_reply
ICB = in_cookie(A),
check:
DiA == gen_digest (ChB, ICB)?
- if OK:
OCB = out_cookie(A),
DiB = gen_digest (ChA, OCB)
(DiB)
<----------------------------------------------- send_challenge_ack
recv_challenge_ack DONE
ICA = in_cookie(B), - else:
check: CLOSE
DiB == gen_digest(ChA, ICA)?
- if OK:
DONE
- else:
CLOSE</pre>
</section>
<section>
<marker id="dflags"/>
<title>Distribution Flags</title>
<p>The following capability flags are defined:</p>
<taglist>
<tag><c>-define(DFLAG_PUBLISHED,16#1).</c></tag>
<item>
<p>The node is to be published and part of the global namespace.</p>
</item>
<tag><c>-define(DFLAG_ATOM_CACHE,16#2).</c></tag>
<item>
<p>The node implements an atom cache (obsolete).</p>
</item>
<tag><c>-define(DFLAG_EXTENDED_REFERENCES,16#4).</c></tag>
<item>
<p>The node implements extended (3 × 32 bits) references. This
is required today. If not present, the connection is refused.</p>
</item>
<tag><c>-define(DFLAG_DIST_MONITOR,16#8).</c></tag>
<item>
<p>The node implements distributed process monitoring.</p>
</item>
<tag><c>-define(DFLAG_FUN_TAGS,16#10).</c></tag>
<item>
<p>The node uses separate tag for funs (lambdas) in the distribution
protocol.</p>
</item>
<tag><c>-define(DFLAG_DIST_MONITOR_NAME,16#20).</c></tag>
<item>
<p>The node implements distributed named process monitoring.</p>
</item>
<tag><c>-define(DFLAG_HIDDEN_ATOM_CACHE,16#40).</c></tag>
<item>
<p>The (hidden) node implements atom cache (obsolete).</p>
</item>
<tag><c>-define(DFLAG_NEW_FUN_TAGS,16#80).</c></tag>
<item>
<p>The node understand new fun tags.</p>
</item>
<tag><c>-define(DFLAG_EXTENDED_PIDS_PORTS,16#100).</c></tag>
<item>
<p>The node can handle extended pids and ports. This is required
today. If not present, the connection is refused.</p>
</item>
<tag><c>-define(DFLAG_EXPORT_PTR_TAG,16#200).</c></tag>
<item>
</item>
<tag><c>-define(DFLAG_BIT_BINARIES,16#400).</c></tag>
<item>
</item>
<tag><c>-define(DFLAG_NEW_FLOATS,16#800).</c></tag>
<item>
<p>The node understands new float format.</p>
</item>
<tag><c>-define(DFLAG_UNICODE_IO,16#1000).</c></tag>
<item>
</item>
<tag><c>-define(DFLAG_DIST_HDR_ATOM_CACHE,16#2000).</c></tag>
<item>
<p>The node implements atom cache in distribution header.</p>
</item>
<tag><c>-define(DFLAG_SMALL_ATOM_TAGS, 16#4000).</c></tag>
<item>
<p>The node understand the <c>SMALL_ATOM_EXT</c> tag.</p>
</item>
<tag><c>-define(DFLAG_UTF8_ATOMS, 16#10000).</c></tag>
<item>
<p>The node understand UTF-8 encoded atoms.</p>
</item>
<tag><c>-define(DFLAG_MAP_TAG, 16#20000).</c></tag>
<item>
<p>The node understand the map tag.</p>
</item>
<tag><c>-define(DFLAG_BIG_CREATION, 16#40000).</c></tag>
<item>
<p>The node understand big node creation.</p>
</item>
<tag><c>-define(DFLAG_SEND_SENDER, 16#80000).</c></tag>
<item>
<p>
Use the <c>SEND_SENDER</c>
<seealso marker="#control_message">control message</seealso>
instead of the <c>SEND</c> control message and use the
<c>SEND_SENDER_TT</c> control message instead
of the <c>SEND_TT</c> control message.
</p>
</item>
<tag><c>-define(DFLAG_BIG_SEQTRACE_LABELS, 16#100000).</c></tag>
<item>
<p>The node understands any term as the seqtrace label.</p>
</item>
<tag><c>-define(DFLAG_EXIT_PAYLOAD, 16#400000).</c></tag>
<item>
<p>Use the <c>PAYLOAD_EXIT</c>, <c>PAYLOAD_EXIT_TT</c>,
<c>PAYLOAD_EXIT2</c>, <c>PAYLOAD_EXIT2_TT</c>
and <c>PAYLOAD_MONITOR_P_EXIT</c>
<seealso marker="#control_message">control message</seealso>s
instead of the non-PAYLOAD variants.</p>
</item>
</taglist>
<p>
There is also function <c>dist_util:strict_order_flags/0</c>
returning all flags (bitwise or:ed together) corresponding to features
that require strict ordering of data over distribution channels.
</p>
</section>
</section>
<section>
<marker id="connected_nodes"/>
<title>Protocol between Connected Nodes</title>
<p>Since ERTS 5.7.2 (OTP R13B) the runtime system passes a distribution flag
in the handshake stage that enables the use of a
<seealso marker="erl_ext_dist#distribution_header">distribution header
</seealso> on all messages passed. Messages passed between nodes have in
this case the following format:</p>
<table align="left">
<row>
<cell align="center">4</cell>
<cell align="center">d</cell>
<cell align="center">n</cell>
<cell align="center">m</cell>
</row>
<row>
<cell align="center"><c>Length</c></cell>
<cell align="center"><c>DistributionHeader</c></cell>
<cell align="center"><c>ControlMessage</c></cell>
<cell align="center"><c>Message</c></cell>
</row>
<tcaption>Format of Messages Passed between Nodes (as from ERTS 5.7.2 (OTP R13B))
</tcaption>
</table>
<taglist>
<tag><c>Length</c></tag>
<item>
<p>Equal to d + n + m.</p>
</item>
<tag><c>DistributionHeader</c></tag>
<item>
<p>
<seealso marker="erl_ext_dist#distribution_header">Distribution header
describing the atom cache and fragmented distribution messages.
</seealso>
</p>
</item>
<tag><c>ControlMessage</c></tag>
<item>
<p>A tuple passed using the external format of Erlang.</p>
</item>
<tag><c>Message</c></tag>
<item>
<p>The message sent to another node using the '!'
or the reason for a EXIT, EXIT2 or DOWN signal using
the external term format.</p>
</item>
</taglist>
<p>Notice that <seealso marker="erl_ext_dist#overall_format">the version
number is omitted from the terms that follow a distribution header
</seealso>.</p>
<p>Nodes with an ERTS version earlier than 5.7.2 (OTP R13B) does not pass the
distribution flag that enables the distribution header. Messages passed
between nodes have in this case the following format:</p>
<table align="left">
<row>
<cell align="center">4</cell>
<cell align="center">1</cell>
<cell align="center">n</cell>
<cell align="center">m</cell>
</row>
<row>
<cell align="center"><c>Length</c></cell>
<cell align="center"><c>Type</c></cell>
<cell align="center"><c>ControlMessage</c></cell>
<cell align="center"><c>Message</c></cell>
</row>
<tcaption>Format of Messages Passed between Nodes (before ERTS 5.7.2 (OTP R13B))
</tcaption>
</table>
<taglist>
<tag><c>Length</c></tag>
<item>
<p>Equal to 1 + n + m.</p>
</item>
<tag><c>Type</c></tag>
<item>
<p>Equal to <c>112</c> (pass through).</p>
</item>
<tag><c>ControlMessage</c></tag>
<item>
<p>A tuple passed using the external format of Erlang.</p>
</item>
<tag><c>Message</c></tag>
<item>
<p>The message sent to another node using the '!' (in external format).
Notice that <c>Message</c> is only passed in combination with a
<c>ControlMessage</c> encoding a send ('!').</p>
</item>
</taglist>
<marker id="control_message"/>
<p>The <c>ControlMessage</c> is a tuple, where the first element indicates
which distributed operation it encodes:</p>
<taglist>
<tag><c>LINK</c></tag>
<item>
<p><c>{1, FromPid, ToPid}</c></p>
</item>
<tag><c>SEND</c></tag>
<item>
<p><c>{2, Unused, ToPid}</c></p>
<p>Followed by <c>Message</c>.</p>
<p><c>Unused</c> is kept for backward compatibility.</p>
</item>
<tag><c>EXIT</c></tag>
<item>
<p><c>{3, FromPid, ToPid, Reason}</c></p>
<p>This signal is sent when a link has been broken</p>
</item>
<tag><c>UNLINK</c></tag>
<item>
<p><c>{4, FromPid, ToPid}</c></p>
</item>
<tag><c>NODE_LINK</c></tag>
<item>
<p><c>{5}</c></p>
</item>
<tag><c>REG_SEND</c></tag>
<item>
<p><c>{6, FromPid, Unused, ToName}</c></p>
<p>Followed by <c>Message</c>.</p>
<p><c>Unused</c> is kept for backward compatibility.</p>
</item>
<tag><c>GROUP_LEADER</c></tag>
<item>
<p><c>{7, FromPid, ToPid}</c></p>
</item>
<tag><c>EXIT2</c></tag>
<item>
<p><c>{8, FromPid, ToPid, Reason}</c></p>
<p>This signal is sent by a call to the erlang:exit/2 bif</p>
</item>
</taglist>
</section>
<section>
<title>New Ctrlmessages for distrvsn = 1 (Erlang/OTP R4)</title>
<taglist>
<tag><c>SEND_TT</c></tag>
<item>
<p><c>{12, Unused, ToPid, TraceToken}</c></p>
<p>Followed by <c>Message</c>.</p>
<p><c>Unused</c> is kept for backward compatibility.</p>
</item>
<tag><c>EXIT_TT</c></tag>
<item>
<p><c>{13, FromPid, ToPid, TraceToken, Reason}</c></p>
</item>
<tag><c>REG_SEND_TT</c></tag>
<item>
<p><c>{16, FromPid, Unused, ToName, TraceToken}</c></p>
<p>Followed by <c>Message</c>.</p>
<p><c>Unused</c> is kept for backward compatibility.</p>
</item>
<tag><c>EXIT2_TT</c></tag>
<item>
<p><c>{18, FromPid, ToPid, TraceToken, Reason}</c></p>
</item>
</taglist>
</section>
<section>
<title>New Ctrlmessages for distrvsn = 2</title>
<p><c>distrvsn</c> 2 was never used.</p>
</section>
<section>
<title>New Ctrlmessages for distrvsn = 3 (Erlang/OTP R5C)</title>
<p>None, but the version number was increased anyway.</p>
</section>
<section>
<title>New Ctrlmessages for distrvsn = 4 (Erlang/OTP R6)</title>
<p>These are only recognized by Erlang nodes, not by hidden nodes.</p>
<taglist>
<tag><c>MONITOR_P</c></tag>
<item>
<p><c>{19, FromPid, ToProc, Ref}</c>, where
<c>FromPid</c> = monitoring process and
<c>ToProc</c> = monitored process pid or name (atom)</p>
</item>
<tag><c>DEMONITOR_P</c></tag>
<item>
<p><c>{20, FromPid, ToProc, Ref}</c>, where
<c>FromPid</c> = monitoring process and
<c>ToProc</c> = monitored process pid or name (atom)</p>
<p>We include <c>FromPid</c> just in case we want to trace this.</p>
</item>
<tag><c>MONITOR_P_EXIT</c></tag>
<item>
<p><c>{21, FromProc, ToPid, Ref, Reason}</c>, where
<c>FromProc</c> = monitored process pid or name (atom),
<c>ToPid</c> = monitoring process, and
<c>Reason</c> = exit reason for the monitored process</p>
</item>
</taglist>
</section>
<section>
<title>New Ctrlmessages for Erlang/OTP 21</title>
<taglist>
<tag><c>SEND_SENDER</c></tag>
<item>
<p><c>{22, FromPid, ToPid}</c></p>
<p>Followed by <c>Message</c>.</p>
<p>
This control message replaces the <c>SEND</c> control
message and will be sent when the distribution flag
<seealso marker="erl_dist_protocol#dflags"><c>DFLAG_SEND_SENDER</c></seealso>
has been negotiated in the connection setup handshake.
</p>
<note><p>
Messages encoded before the connection has
been set up may still use the <c>SEND</c> control
message. However, once a <c>SEND_SENDER</c> or <c>SEND_SENDER_TT</c>
control message has been sent, no more <c>SEND</c>
control messages will be sent in the same direction
on the connection.
</p></note>
</item>
<tag><c>SEND_SENDER_TT</c></tag>
<item>
<p><c>{23, FromPid, ToPid, TraceToken}</c></p>
<p>Followed by <c>Message</c>.</p>
<p>
This control message replaces the <c>SEND_TT</c> control
message and will be sent when the distribution flag
<seealso marker="erl_dist_protocol#dflags"><c>DFLAG_SEND_SENDER</c></seealso>
has been negotiated in the connection setup handshake.
</p>
<note><p>
Messages encoded before the connection has
been set up may still use the <c>SEND_TT</c> control
message. However, once a <c>SEND_SENDER</c> or <c>SEND_SENDER_TT</c>
control message has been sent, no more <c>SEND_TT</c>
control messages will be sent in the same direction
on the connection.
</p></note>
</item>
</taglist>
</section>
<section>
<title>New Ctrlmessages for Erlang/OTP 22</title>
<note><p>
Messages encoded before the connection has
been set up may still use the non-PAYLOAD variant.
However, once a PAYLOAD control message has been sent,
no more non-PAYLOAD control messages will be sent in
the same direction on the connection.
</p></note>
<taglist>
<tag><c>PAYLOAD_EXIT</c></tag>
<item>
<p><c>{24, FromPid, ToPid}</c></p>
<p>Followed by <c>Reason</c>.</p>
<p>
This control message replaces the <c>EXIT</c> control
message and will be sent when the distribution flag
<seealso marker="erl_dist_protocol#dflags"><c>DFLAG_EXIT_PAYLOAD</c></seealso>
has been negotiated in the connection setup handshake.
</p>
</item>
<tag><c>PAYLOAD_EXIT_TT</c></tag>
<item>
<p><c>{25, FromPid, ToPid}</c></p>
<p>Followed by <c>Reason</c>.</p>
<p>
This control message replaces the <c>EXIT_TT</c> control
message and will be sent when the distribution flag
<seealso marker="erl_dist_protocol#dflags"><c>DFLAG_EXIT_PAYLOAD</c></seealso>
has been negotiated in the connection setup handshake.
</p>
</item>
<tag><c>PAYLOAD_EXIT2</c></tag>
<item>
<p><c>{26, FromPid, ToPid}</c></p>
<p>Followed by <c>Reason</c>.</p>
<p>
This control message replaces the <c>EXIT2</c> control
message and will be sent when the distribution flag
<seealso marker="erl_dist_protocol#dflags"><c>DFLAG_EXIT_PAYLOAD</c></seealso>
has been negotiated in the connection setup handshake.
</p>
</item>
<tag><c>PAYLOAD_EXIT2_TT</c></tag>
<item>
<p><c>{27, FromPid, ToPid}</c></p>
<p>Followed by <c>Reason</c>.</p>
<p>
This control message replaces the <c>EXIT2_TT</c> control
message and will be sent when the distribution flag
<seealso marker="erl_dist_protocol#dflags"><c>DFLAG_EXIT_PAYLOAD</c></seealso>
has been negotiated in the connection setup handshake.
</p>
</item>
<tag><c>PAYLOAD_MONITOR_P_EXIT</c></tag>
<item>
<p><c>{28, FromPid, ToPid, Ref}</c></p>
<p>Followed by <c>Reason</c>.</p>
<p>
This control message replaces the <c>MONITOR_P_EXIT</c> control
message and will be sent when the distribution flag
<seealso marker="erl_dist_protocol#dflags"><c>DFLAG_EXIT_PAYLOAD</c></seealso>
has been negotiated in the connection setup handshake.
</p>
</item>
</taglist>
</section>
</chapter>