Erlang logo
User's Guide
Reference Manual
Release Notes

Erlang Run-Time System Application (ERTS)
Reference Manual
Version 5.8.4

Expand All
Contract All

Table of Contents





Erlang Port Mapper Daemon

epmd [-d|-debug] [DbgExtra...] [-port No] [-daemon] [-relaxed_command_check]

Starts the port mapper daemon

epmd [-d|-debug] [-port No] [-names|-kill|-stop Name]

Communicates with a running port mapper daemon


This daemon acts as a name server on all hosts involved in distributed Erlang computations. When an Erlang node starts, the node has a name and it obtains an address from the host OS kernel. The name and the address are sent to the epmd daemon running on the local host. In a TCP/IP environment, the address consists of the IP address and a port number. The name of the node is an atom on the form of Name@Node. The job of the epmd daemon is to keep track of which node name listens on which address. Hence, epmd map symbolic node names to machine addresses.

The TCP/IP epmd daemon actually only keeps track of the Name (first) part of an Erlang node name, the Host part (whatever is after the @ is implicit in the node name where the epmd daemon was actually contacted, as is the IP address where the Erlang node can be reached. Consistent and correct TCP naming services are therefore required for an Erlang network to function correctly.

Starting the port mapper daemon

The daemon is started automatically by the erl command if the node is to be distributed and there is no running instance present. If automatically launched, environment variables has to be used to alter the behavior of the daemon. See the Environment variables section below.

If the -daemon argument is not given, the epmd runs as a normal program with the controlling terminal of the shell in which it is started. Normally, it should run as a daemon.

Regular start-up options are described in the Regular options section below.

The DbgExtra options are described in the DbgExtra options section below.

Communicating with a running port mapper daemon

Communicating with the running epmd daemon by means of the epmd program is done primarily for debugging purposes.

The different queries are described in the Interactive options section below.

Regular options

These options are available when starting the actual name server. The name server is normally started automatically by the erl command (if not already available), but it can also be started at i.e. system start-up.

-address List

Let this instance of epmd listen only on the comma-separated list of IP addresses and on the loopback address (which is implicitely added to the list if it has not been specified). This can also be set using the ERL_EPMD_ADDRESS environment variable, see the section Environment variables below.

-port No

Let this instance of epmd listen to another TCP port than default 4369. This can also be set using the ERL_EPMD_PORT environment variable, see the section Environment variables below

-d | -debug

Enable debug output. The more -d flags given, the more debug output you will get (to a certain limit). This option is most useful when the epmd daemon is not started as a daemon.


Start epmd detached from the controlling terminal. Logging will end up in syslog when available and correctly configured. If the epmd daemon is started at boot, this option should definitely be used. It is also used when the erl command automatically starts epmd.


Start the epmd program with relaxed command checking (mostly for backward compatibility). This affects the following:

  • With relaxed command checking, the epmd daemon can be killed from the localhost with i.e. epmd -kill even if there are active nodes registered. Normally only daemons with an empty node database can be killed with the epmd -kill command.

  • The epmd -stop command (and the corresponding messages to epmd, as can be given using erl_interface/ei) is normally always ignored, as it opens up for strange situation when two nodes of the same name can be alive at the same time. A node unregisters itself by just closing the connection to epmd, why the stop command was only intended for use in debugging situations.

    With relaxed command checking enabled, you can forcibly unregister live nodes.

Relaxed command checking can also be enabled by setting the environment variable ERL_EPMD_RELAXED_COMMAND_CHECK prior to starting epmd.

Only use relaxed command checking on systems with very limited interactive usage.

DbgExtra options

These options are purely for debugging and testing epmd clients, they should not be used in normal operation.

-packet_timeout Seconds

Set the number of seconds a connection can be inactive before epmd times out and closes the connection (default 60).

-delay_accept Seconds

To simulate a busy server you can insert a delay between epmd gets notified about that a new connection is requested and when the connections gets accepted.

-delay_write Seconds

Also a simulation of a busy server. Inserts a delay before a reply is sent.

Interactive options

These options make epmd run as an interactive command displaying the results of sending queries ta an already running instance of epmd. The epmd contacted is always on the local node, but the -port option can be used to select between instances if several are running using different port on the host.

-port No

Contacts the epmd listening on the given TCP port number (default 4369). This can also be set using the ERL_EPMD_PORT environment variable, see the section Environment variables below


List names registered with the currently running epmd


Kill the currently running epmd.

Killing the running epmd is only allowed if epmd -names show an empty database or -relaxed_command_check was given when the running instance of epmd was started. Note that -relaxed_command_check is given when starting the daemon that is to accept killing when it has live nodes registered. When running epmd interactively, -relaxed_command_check has no effect. A daemon that is started without relaxed command checking has to be killed using i.e. signals or some other OS specific method if it has active clients registered.

-stop Name

Forcibly unregister a live node from epmd's database

This command can only be used when contacting epmd instances started with the -relaxed_command_check flag. Note that relaxed command checking has to be enabled for the epmd daemon contacted, When running epmd interactively, -relaxed_command_check has no effect.

Environment variables


This environment variable may be set to a comma-separated list of IP addresses, in which case the epmd daemon will listen only on the specified address(es) and on the loopback address (which is implicitely added to the list if it has not been specified). The default behaviour is to listen on all available IP addresses.


This environment variable can contain the port number epmd will use. The default port will work fine in most cases. A different port can be specified to allow several instances of epmd, representing independent clusters of nodes, to co-exist on the same host. All nodes in a cluster must use the same epmd port number.


If set prior to start, the epmd daemon will behave as if the -relaxed_command_check option was given at start-up. If consequently setting this option before starting the Erlang virtual machine, the automatically started epmd will accept the -kill and -stop commands without restrictions.


On some operating systems syslog will be used for error reporting when epmd runs as an daemon. To enable the error logging you have to edit /etc/syslog.conf file and add an entry


where <TABs> are at least one real tab character. Spaces will silently be ignored.

Access restrictions

The epmd daemon accepts messages from both localhost and remote hosts. However, only the query commands are answered (and acted upon) if the query comes from a remote host. It is always an error to try to register a nodename if the client is not a process located on the same host as the epmd instance is running on, why such requests are considered hostile and the connection is immediately closed.

The queries accepted from remote nodes are:

  • Port queries - i.e. on which port does the node with a given name listen

  • Name listing - i.e. give a list of all names registered on the host

To restrict access further, firewall software has to be used.