Current File : //usr/share/doc/openvpn/openvpn.8.html
<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
<title>openvpn</title>
<style type="text/css">
/*
:Author: David Goodger (goodger@python.org)
:Id: $Id: html4css1.css 7952 2016-07-26 18:15:59Z milde $
:Copyright: This stylesheet has been placed in the public domain.
Default cascading style sheet for the HTML output of Docutils.
See http://docutils.sf.net/docs/howto/html-stylesheets.html for how to
customize this style sheet.
*/
/* used to remove borders from tables and images */
.borderless, table.borderless td, table.borderless th {
border: 0 }
table.borderless td, table.borderless th {
/* Override padding for "table.docutils td" with "! important".
The right padding separates the table cells. */
padding: 0 0.5em 0 0 ! important }
.first {
/* Override more specific margin styles with "! important". */
margin-top: 0 ! important }
.last, .with-subtitle {
margin-bottom: 0 ! important }
.hidden {
display: none }
.subscript {
vertical-align: sub;
font-size: smaller }
.superscript {
vertical-align: super;
font-size: smaller }
a.toc-backref {
text-decoration: none ;
color: black }
blockquote.epigraph {
margin: 2em 5em ; }
dl.docutils dd {
margin-bottom: 0.5em }
object[type="image/svg+xml"], object[type="application/x-shockwave-flash"] {
overflow: hidden;
}
/* Uncomment (and remove this text!) to get bold-faced definition list terms
dl.docutils dt {
font-weight: bold }
*/
div.abstract {
margin: 2em 5em }
div.abstract p.topic-title {
font-weight: bold ;
text-align: center }
div.admonition, div.attention, div.caution, div.danger, div.error,
div.hint, div.important, div.note, div.tip, div.warning {
margin: 2em ;
border: medium outset ;
padding: 1em }
div.admonition p.admonition-title, div.hint p.admonition-title,
div.important p.admonition-title, div.note p.admonition-title,
div.tip p.admonition-title {
font-weight: bold ;
font-family: sans-serif }
div.attention p.admonition-title, div.caution p.admonition-title,
div.danger p.admonition-title, div.error p.admonition-title,
div.warning p.admonition-title, .code .error {
color: red ;
font-weight: bold ;
font-family: sans-serif }
/* Uncomment (and remove this text!) to get reduced vertical space in
compound paragraphs.
div.compound .compound-first, div.compound .compound-middle {
margin-bottom: 0.5em }
div.compound .compound-last, div.compound .compound-middle {
margin-top: 0.5em }
*/
div.dedication {
margin: 2em 5em ;
text-align: center ;
font-style: italic }
div.dedication p.topic-title {
font-weight: bold ;
font-style: normal }
div.figure {
margin-left: 2em ;
margin-right: 2em }
div.footer, div.header {
clear: both;
font-size: smaller }
div.line-block {
display: block ;
margin-top: 1em ;
margin-bottom: 1em }
div.line-block div.line-block {
margin-top: 0 ;
margin-bottom: 0 ;
margin-left: 1.5em }
div.sidebar {
margin: 0 0 0.5em 1em ;
border: medium outset ;
padding: 1em ;
background-color: #ffffee ;
width: 40% ;
float: right ;
clear: right }
div.sidebar p.rubric {
font-family: sans-serif ;
font-size: medium }
div.system-messages {
margin: 5em }
div.system-messages h1 {
color: red }
div.system-message {
border: medium outset ;
padding: 1em }
div.system-message p.system-message-title {
color: red ;
font-weight: bold }
div.topic {
margin: 2em }
h1.section-subtitle, h2.section-subtitle, h3.section-subtitle,
h4.section-subtitle, h5.section-subtitle, h6.section-subtitle {
margin-top: 0.4em }
h1.title {
text-align: center }
h2.subtitle {
text-align: center }
hr.docutils {
width: 75% }
img.align-left, .figure.align-left, object.align-left, table.align-left {
clear: left ;
float: left ;
margin-right: 1em }
img.align-right, .figure.align-right, object.align-right, table.align-right {
clear: right ;
float: right ;
margin-left: 1em }
img.align-center, .figure.align-center, object.align-center {
display: block;
margin-left: auto;
margin-right: auto;
}
table.align-center {
margin-left: auto;
margin-right: auto;
}
.align-left {
text-align: left }
.align-center {
clear: both ;
text-align: center }
.align-right {
text-align: right }
/* reset inner alignment in figures */
div.align-right {
text-align: inherit }
/* div.align-center * { */
/* text-align: left } */
.align-top {
vertical-align: top }
.align-middle {
vertical-align: middle }
.align-bottom {
vertical-align: bottom }
ol.simple, ul.simple {
margin-bottom: 1em }
ol.arabic {
list-style: decimal }
ol.loweralpha {
list-style: lower-alpha }
ol.upperalpha {
list-style: upper-alpha }
ol.lowerroman {
list-style: lower-roman }
ol.upperroman {
list-style: upper-roman }
p.attribution {
text-align: right ;
margin-left: 50% }
p.caption {
font-style: italic }
p.credits {
font-style: italic ;
font-size: smaller }
p.label {
white-space: nowrap }
p.rubric {
font-weight: bold ;
font-size: larger ;
color: maroon ;
text-align: center }
p.sidebar-title {
font-family: sans-serif ;
font-weight: bold ;
font-size: larger }
p.sidebar-subtitle {
font-family: sans-serif ;
font-weight: bold }
p.topic-title {
font-weight: bold }
pre.address {
margin-bottom: 0 ;
margin-top: 0 ;
font: inherit }
pre.literal-block, pre.doctest-block, pre.math, pre.code {
margin-left: 2em ;
margin-right: 2em }
pre.code .ln { color: grey; } /* line numbers */
pre.code, code { background-color: #eeeeee }
pre.code .comment, code .comment { color: #5C6576 }
pre.code .keyword, code .keyword { color: #3B0D06; font-weight: bold }
pre.code .literal.string, code .literal.string { color: #0C5404 }
pre.code .name.builtin, code .name.builtin { color: #352B84 }
pre.code .deleted, code .deleted { background-color: #DEB0A1}
pre.code .inserted, code .inserted { background-color: #A3D289}
span.classifier {
font-family: sans-serif ;
font-style: oblique }
span.classifier-delimiter {
font-family: sans-serif ;
font-weight: bold }
span.interpreted {
font-family: sans-serif }
span.option {
white-space: nowrap }
span.pre {
white-space: pre }
span.problematic {
color: red }
span.section-subtitle {
/* font-size relative to parent (h1..h6 element) */
font-size: 80% }
table.citation {
border-left: solid 1px gray;
margin-left: 1px }
table.docinfo {
margin: 2em 4em }
table.docutils {
margin-top: 0.5em ;
margin-bottom: 0.5em }
table.footnote {
border-left: solid 1px black;
margin-left: 1px }
table.docutils td, table.docutils th,
table.docinfo td, table.docinfo th {
padding-left: 0.5em ;
padding-right: 0.5em ;
vertical-align: top }
table.docutils th.field-name, table.docinfo th.docinfo-name {
font-weight: bold ;
text-align: left ;
white-space: nowrap ;
padding-left: 0 }
/* "booktabs" style (no vertical lines) */
table.docutils.booktabs {
border: 0px;
border-top: 2px solid;
border-bottom: 2px solid;
border-collapse: collapse;
}
table.docutils.booktabs * {
border: 0px;
}
table.docutils.booktabs th {
border-bottom: thin solid;
text-align: left;
}
h1 tt.docutils, h2 tt.docutils, h3 tt.docutils,
h4 tt.docutils, h5 tt.docutils, h6 tt.docutils {
font-size: 100% }
ul.auto-toc {
list-style-type: none }
</style>
</head>
<body>
<div class="document" id="openvpn">
<h1 class="title">openvpn</h1>
<h2 class="subtitle" id="secure-ip-tunnel-daemon">Secure IP tunnel daemon</h2>
<table class="docinfo" frame="void" rules="none">
<col class="docinfo-name" />
<col class="docinfo-content" />
<tbody valign="top">
<tr class="manual-section field"><th class="docinfo-name">Manual section:</th><td class="field-body">8</td>
</tr>
<tr class="manual-group field"><th class="docinfo-name">Manual group:</th><td class="field-body">System Manager's Manual</td>
</tr>
</tbody>
</table>
<div class="section" id="synopsis">
<h1>SYNOPSIS</h1>
<div class="line-block">
<div class="line"><tt class="docutils literal">openvpn</tt> [ options ... ]</div>
<div class="line"><tt class="docutils literal">openvpn</tt> <tt class="docutils literal"><span class="pre">--help</span></tt></div>
</div>
</div>
<div class="section" id="introduction">
<h1>INTRODUCTION</h1>
<p>OpenVPN is an open source VPN daemon by James Yonan. Because OpenVPN
tries to be a universal VPN tool offering a great deal of flexibility,
there are a lot of options on this manual page. If you're new to
OpenVPN, you might want to skip ahead to the examples section where you
will see how to construct simple VPNs on the command line without even
needing a configuration file.</p>
<p>Also note that there's more documentation and examples on the OpenVPN
web site: <a class="reference external" href="https://openvpn.net/">https://openvpn.net/</a></p>
<p>And if you would like to see a shorter version of this manual, see the
openvpn usage message which can be obtained by running <strong>openvpn</strong>
without any parameters.</p>
</div>
<div class="section" id="description">
<h1>DESCRIPTION</h1>
<p>OpenVPN is a robust and highly flexible VPN daemon. OpenVPN supports
SSL/TLS security, ethernet bridging, TCP or UDP tunnel transport through
proxies or NAT, support for dynamic IP addresses and DHCP, scalability
to hundreds or thousands of users, and portability to most major OS
platforms.</p>
<p>OpenVPN is tightly bound to the OpenSSL library, and derives much of its
crypto capabilities from it.</p>
<p>OpenVPN supports conventional encryption using a pre-shared secret key
<strong>(Static Key mode)</strong> or public key security <strong>(SSL/TLS mode)</strong> using
client & server certificates. OpenVPN also supports non-encrypted
TCP/UDP tunnels.</p>
<p>OpenVPN is designed to work with the <strong>TUN/TAP</strong> virtual networking
interface that exists on most platforms.</p>
<p>Overall, OpenVPN aims to offer many of the key features of IPSec but
with a relatively lightweight footprint.</p>
</div>
<div class="section" id="options">
<h1>OPTIONS</h1>
<p>OpenVPN allows any option to be placed either on the command line or in
a configuration file. Though all command line options are preceded by a
double-leading-dash ("--"), this prefix can be removed when an option is
placed in a configuration file.</p>
<div class="section" id="generic-options">
<h2>Generic Options</h2>
<p>This section covers generic options which are accessible regardless of
which mode OpenVPN is configured as.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group">
<kbd><span class="option">--help</span></kbd></td>
<td>Show options.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--auth-nocache</span></kbd></td>
<td><p class="first">Don't cache <tt class="docutils literal"><span class="pre">--askpass</span></tt> or <tt class="docutils literal"><span class="pre">--auth-user-pass</span></tt> username/passwords in
virtual memory.</p>
<p>If specified, this directive will cause OpenVPN to immediately forget
username/password inputs after they are used. As a result, when OpenVPN
needs a username/password, it will prompt for input from stdin, which
may be multiple times during the duration of an OpenVPN session.</p>
<p>When using <tt class="docutils literal"><span class="pre">--auth-nocache</span></tt> in combination with a user/password file
and <tt class="docutils literal"><span class="pre">--chroot</span></tt> or <tt class="docutils literal"><span class="pre">--daemon</span></tt>, make sure to use an absolute path.</p>
<p class="last">This directive does not affect the <tt class="docutils literal"><span class="pre">--http-proxy</span></tt> username/password.
It is always cached.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--cd <var>dir</var></span></kbd></td>
<td><p class="first">Change directory to <tt class="docutils literal">dir</tt> prior to reading any files such as
configuration files, key files, scripts, etc. <tt class="docutils literal">dir</tt> should be an
absolute path, with a leading "/", and without any references to the
current directory such as <code>.</code> or <code>..</code>.</p>
<p class="last">This option is useful when you are running OpenVPN in <tt class="docutils literal"><span class="pre">--daemon</span></tt> mode,
and you want to consolidate all of your OpenVPN control files in one
location.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--chroot <var>dir</var></span></kbd></td>
<td><p class="first">Chroot to <tt class="docutils literal">dir</tt> after initialization. <tt class="docutils literal"><span class="pre">--chroot</span></tt> essentially
redefines <tt class="docutils literal">dir</tt> as being the top level directory tree (/). OpenVPN
will therefore be unable to access any files outside this tree. This can
be desirable from a security standpoint.</p>
<p>Since the chroot operation is delayed until after initialization, most
OpenVPN options that reference files will operate in a pre-chroot
context.</p>
<p>In many cases, the <tt class="docutils literal">dir</tt> parameter can point to an empty directory,
however complications can result when scripts or restarts are executed
after the chroot operation.</p>
<p class="last">Note: The SSL library will probably need /dev/urandom to be available
inside the chroot directory <tt class="docutils literal">dir</tt>. This is because SSL libraries
occasionally need to collect fresh randomness. Newer linux kernels and some
BSDs implement a getrandom() or getentropy() syscall that removes the
need for /dev/urandom to be available.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--compat-mode <var>version</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">This option provides a convenient way to alter the defaults of OpenVPN
to be more compatible with the version <tt class="docutils literal">version</tt> specified. All of
the changes this option applies can also be achieved using individual
configuration options.</p>
<p>The version specified with this option is the version of OpenVPN peer
OpenVPN should try to be compatible with. In general OpenVPN should be
compatible with the last two previous version without this option. E.g.
OpenVPN 2.6.0 should be compatible with 2.5.x and 2.4.x without this option.
However, there might be some edge cases that still require this option even
in these cases.</p>
<p>Note: Using this option reverts defaults to no longer recommended
values and should be avoided if possible.</p>
<p>The following table details what defaults are changed depending on the
version specified.</p>
<ul class="simple">
<li>2.5.x or lower: <tt class="docutils literal"><span class="pre">--allow-compression</span> asym</tt> is automatically added
to the configuration if no other compression options are present.</li>
<li>2.4.x or lower: The cipher in <tt class="docutils literal"><span class="pre">--cipher</span></tt> is appended to
<tt class="docutils literal"><span class="pre">--data-ciphers</span></tt>.</li>
<li>2.3.x or lower: <tt class="docutils literal"><span class="pre">--data-cipher-fallback</span></tt> is automatically added with
the same cipher as <tt class="docutils literal"><span class="pre">--cipher</span></tt>.</li>
<li>2.3.6 or lower: <tt class="docutils literal"><span class="pre">--tls-version-min</span> 1.0</tt> is added to the configuration
when <tt class="docutils literal"><span class="pre">--tls-version-min</span></tt> is not explicitly set.</li>
</ul>
<p class="last">If not required, this is option should be avoided. Setting this option can
lower security or disable features like data-channel offloading.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--config <var>file</var></span></kbd></td>
<td><p class="first">Load additional config options from <tt class="docutils literal">file</tt> where each line corresponds
to one command line option, but with the leading <code>--</code> removed.</p>
<p>If <tt class="docutils literal"><span class="pre">--config</span> file</tt> is the only option to the openvpn command, the
<tt class="docutils literal"><span class="pre">--config</span></tt> can be removed, and the command can be given as <tt class="docutils literal">openvpn
file</tt></p>
<p>Note that configuration files can be nested to a reasonable depth.</p>
<p>Double quotation or single quotation characters ("", '') can be used to
enclose single parameters containing whitespace, and "#" or ";"
characters in the first column can be used to denote comments.</p>
<p>Note that OpenVPN 2.0 and higher performs backslash-based shell escaping
for characters not in single quotations, so the following mappings
should be observed:</p>
<pre class="literal-block">
\\ Maps to a single backslash character (\).
\" Pass a literal doublequote character ("), don't
interpret it as enclosing a parameter.
\[SPACE] Pass a literal space or tab character, don't
interpret it as a parameter delimiter.
</pre>
<p>For example on Windows, use double backslashes to represent pathnames:</p>
<pre class="literal-block">
secret "c:\\OpenVPN\\secret.key"
</pre>
<p>For examples of configuration files, see
<a class="reference external" href="https://openvpn.net/community-resources/how-to/">https://openvpn.net/community-resources/how-to/</a></p>
<p>Here is an example configuration file:</p>
<pre class="last literal-block">
#
# Sample OpenVPN configuration file for
# using a pre-shared static key.
#
# '#' or ';' may be used to delimit comments.
# Use a dynamic tun device.
dev tun
# Our remote peer
remote mypeer.mydomain
# 10.1.0.1 is our local VPN endpoint
# 10.1.0.2 is our remote VPN endpoint
ifconfig 10.1.0.1 10.1.0.2
# Our pre-shared static key
secret static.key
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--daemon <var>progname</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Become a daemon after all initialization functions are completed.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
daemon
daemon progname
</pre>
<p>This option will cause all message and error output to be sent to the syslog
file (such as <code>/var/log/messages</code>), except for the output of
scripts and ifconfig commands, which will go to <code>/dev/null</code> unless
otherwise redirected. The syslog redirection occurs immediately at the
point that <tt class="docutils literal"><span class="pre">--daemon</span></tt> is parsed on the command line even though the
daemonization point occurs later. If one of the <tt class="docutils literal"><span class="pre">--log</span></tt> options is
present, it will supersede syslog redirection.</p>
<p>The optional <tt class="docutils literal">progname</tt> parameter will cause OpenVPN to report its
program name to the system logger as <tt class="docutils literal">progname</tt>. This can be useful in
linking OpenVPN messages in the syslog file with specific tunnels. When
unspecified, <tt class="docutils literal">progname</tt> defaults to <code>openvpn</code>.</p>
<p>When OpenVPN is run with the <tt class="docutils literal"><span class="pre">--daemon</span></tt> option, it will try to delay
daemonization until the majority of initialization functions which are
capable of generating fatal errors are complete. This means that
initialization scripts can test the return status of the openvpn command
for a fairly reliable indication of whether the command has correctly
initialized and entered the packet forwarding event loop.</p>
<p>In OpenVPN, the vast majority of errors which occur after initialization
are non-fatal.</p>
<p>Note: as soon as OpenVPN has daemonized, it can not ask for usernames,
passwords, or key pass phrases anymore. This has certain consequences,
namely that using a password-protected private key will fail unless the
<tt class="docutils literal"><span class="pre">--askpass</span></tt> option is used to tell OpenVPN to ask for the pass phrase
(this requirement is new in v2.3.7, and is a consequence of calling
daemon() before initializing the crypto layer).</p>
<p class="last">Further, using <tt class="docutils literal"><span class="pre">--daemon</span></tt> together with <tt class="docutils literal"><span class="pre">--auth-user-pass</span></tt> (entered
on console) and <tt class="docutils literal"><span class="pre">--auth-nocache</span></tt> will fail as soon as key
renegotiation (and reauthentication) occurs.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--disable-dco</span></kbd></td>
<td><p class="first">Disable "data channel offload" (DCO).</p>
<p>On Linux don't use the ovpn-dco device driver, but rather rely on the
legacy tun module.</p>
<p class="last">You may want to use this option if your server needs to allow clients
older than version 2.4 to connect.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--disable-occ</span></kbd></td>
<td><p class="first"><strong>DEPRECATED</strong> Disable "options consistency check" (OCC) in configurations
that do not use TLS.</p>
<p>Don't output a warning message if option inconsistencies are detected
between peers. An example of an option inconsistency would be where one
peer uses <tt class="docutils literal"><span class="pre">--dev</span> tun</tt> while the other peer uses <tt class="docutils literal"><span class="pre">--dev</span> tap</tt>.</p>
<p class="last">Use of this option is discouraged, but is provided as a temporary fix in
situations where a recent version of OpenVPN must connect to an old
version.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--engine <var>engine-name</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Enable OpenSSL hardware-based crypto engine functionality.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
engine
engine engine-name
</pre>
<p class="last">If <tt class="docutils literal"><span class="pre">engine-name</span></tt> is specified, use a specific crypto engine. Use the
<tt class="docutils literal"><span class="pre">--show-engines</span></tt> standalone option to list the crypto engines which
are supported by OpenSSL.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--fast-io</span></kbd></td>
<td><p class="first">(Experimental) Optimize TUN/TAP/UDP I/O writes by avoiding a call to
poll/epoll/select prior to the write operation. The purpose of such a
call would normally be to block until the device or socket is ready to
accept the write. Such blocking is unnecessary on some platforms which
don't support write blocking on UDP sockets or TUN/TAP devices. In such
cases, one can optimize the event loop by avoiding the poll/epoll/select
call, improving CPU efficiency by 5% to 10%.</p>
<p class="last">This option can only be used on non-Windows systems, when <tt class="docutils literal"><span class="pre">--proto</span>
udp</tt> is specified, and when <tt class="docutils literal"><span class="pre">--shaper</span></tt> is <em>NOT</em> specified.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--group <var>group</var></span></kbd></td>
<td>Similar to the <tt class="docutils literal"><span class="pre">--user</span></tt> option, this option changes the group ID of
the OpenVPN process to <tt class="docutils literal">group</tt> after initialization.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ignore-unknown-option <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Valid syntax:</p>
<pre class="literal-block">
ignore-unknown-options opt1 opt2 opt3 ... optN
</pre>
<p>When one of options <tt class="docutils literal">opt1 ... optN</tt> is encountered in the configuration
file the configuration file parsing does not fail if this OpenVPN version
does not support the option. Multiple <tt class="docutils literal"><span class="pre">--ignore-unknown-option</span></tt> options
can be given to support a larger number of options to ignore.</p>
<p>This option should be used with caution, as there are good security
reasons for having OpenVPN fail if it detects problems in a config file.
Having said that, there are valid reasons for wanting new software
features to gracefully degrade when encountered by older software
versions.</p>
<p class="last"><tt class="docutils literal"><span class="pre">--ignore-unknown-option</span></tt> is available since OpenVPN 2.3.3.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--iproute <var>cmd</var></span></kbd></td>
<td>Set alternate command to execute instead of default <tt class="docutils literal">iproute2</tt> command.
May be used in order to execute OpenVPN in unprivileged environment.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--keying-material-exporter <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Save Exported Keying Material [RFC5705] of <tt class="docutils literal">len</tt> bytes (must be between 16
and 4095 bytes) using <tt class="docutils literal">label</tt> in environment
(<code>exported_keying_material</code>) for use by plugins in
<code>OPENVPN_PLUGIN_TLS_FINAL</code> callback.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
keying-material-exporter label len
</pre>
<p class="last">Note that exporter <tt class="docutils literal">labels</tt> have the potential to collide with existing
PRF labels. In order to prevent this, labels <em>MUST</em> begin with
<code>EXPORTER</code>.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--mlock</span></kbd></td>
<td><p class="first">Disable paging by calling the POSIX mlockall function. Requires that
OpenVPN be initially run as root (though OpenVPN can subsequently
downgrade its UID using the <tt class="docutils literal"><span class="pre">--user</span></tt> option).</p>
<p>Using this option ensures that key material and tunnel data are never
written to disk due to virtual memory paging operations which occur
under most modern operating systems. It ensures that even if an attacker
was able to crack the box running OpenVPN, he would not be able to scan
the system swap file to recover previously used ephemeral keys, which
are used for a period of time governed by the <tt class="docutils literal"><span class="pre">--reneg</span></tt> options (see
below), then are discarded.</p>
<p>The downside of using <tt class="docutils literal"><span class="pre">--mlock</span></tt> is that it will reduce the amount of
physical memory available to other applications.</p>
<p>The limit on how much memory can be locked and how that limit
is enforced are OS-dependent. On Linux the default limit that an
unprivileged process may lock (RLIMIT_MEMLOCK) is low, and if
privileges are dropped later, future memory allocations will very
likely fail. The limit can be increased using ulimit or systemd
directives depending on how OpenVPN is started.</p>
<p class="last">If the platform has the getrlimit(2) system call, OpenVPN will check
for the amount of mlock-able memory before calling mlockall(2), and
tries to increase the limit to 100 MB if less than this is configured.
100 Mb is somewhat arbitrary - it is enough for a moderately-sized
OpenVPN deployment, but the memory usage might go beyond that if the
number of concurrent clients is high.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--nice <var>n</var></span></kbd></td>
<td>Change process priority after initialization (<tt class="docutils literal">n</tt> greater than 0 is
lower priority, <tt class="docutils literal">n</tt> less than zero is higher priority).</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--persist-key</span></kbd></td>
<td><p class="first">Don't re-read key files across <code>SIGUSR1</code> or <tt class="docutils literal"><span class="pre">--ping-restart</span></tt>.</p>
<p>This option can be combined with <tt class="docutils literal"><span class="pre">--user</span></tt> to allow restarts
triggered by the <code>SIGUSR1</code> signal. Normally if you drop root
privileges in OpenVPN, the daemon cannot be restarted since it will now
be unable to re-read protected key files.</p>
<p class="last">This option solves the problem by persisting keys across <code>SIGUSR1</code>
resets, so they don't need to be re-read.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--providers <var>providers</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Load the list of (OpenSSL) providers. This is mainly useful for using an
external provider for key management like tpm2-openssl or to load the
legacy provider with</p>
<pre class="literal-block">
--providers legacy default
</pre>
<p class="last">Behaviour of changing this option between <code>SIGHUP</code> might not be well behaving.
If you need to change/add/remove this option, fully restart OpenVPN.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--remap-usr1 <var>signal</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Control whether internally or externally generated <code>SIGUSR1</code> signals
are remapped to <code>SIGHUP</code> (restart without persisting state) or
<code>SIGTERM</code> (exit).</p>
<p class="last"><tt class="docutils literal">signal</tt> can be set to <code>SIGHUP</code> or <code>SIGTERM</code>. By default,
no remapping occurs.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--script-security <var>level</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">This directive offers policy-level control over OpenVPN's usage of
external programs and scripts. Lower <tt class="docutils literal">level</tt> values are more
restrictive, higher values are more permissive. Settings for <tt class="docutils literal">level</tt>:</p>
<dl class="docutils">
<dt><code>0</code></dt>
<dd>Strictly no calling of external programs.</dd>
<dt><code>1</code></dt>
<dd>(Default) Only call built-in executables such as ifconfig,
ip, route, or netsh.</dd>
<dt><code>2</code></dt>
<dd>Allow calling of built-in executables and user-defined
scripts.</dd>
<dt><code>3</code></dt>
<dd>Allow passwords to be passed to scripts via environmental
variables (potentially unsafe).</dd>
</dl>
<p>OpenVPN releases before v2.3 also supported a <tt class="docutils literal">method</tt> flag which
indicated how OpenVPN should call external commands and scripts. This
could be either <code>execve</code> or <code>system</code>. As of OpenVPN 2.3, this
flag is no longer accepted. In most *nix environments the execve()
approach has been used without any issues.</p>
<p>Some directives such as <tt class="docutils literal"><span class="pre">--up</span></tt> allow options to be passed to the
external script. In these cases make sure the script name does not
contain any spaces or the configuration parser will choke because it
can't determine where the script name ends and script options start.</p>
<p>To run scripts in Windows in earlier OpenVPN versions you needed to
either add a full path to the script interpreter which can parse the
script or use the <tt class="docutils literal">system</tt> flag to run these scripts. As of OpenVPN
2.3 it is now a strict requirement to have full path to the script
interpreter when running non-executables files. This is not needed for
executable files, such as .exe, .com, .bat or .cmd files. For example,
if you have a Visual Basic script, you must use this syntax now:</p>
<pre class="literal-block">
--up 'C:\\Windows\\System32\\wscript.exe C:\\Program\ Files\\OpenVPN\\config\\my-up-script.vbs'
</pre>
<p>Please note the single quote marks and the escaping of the backslashes
(\) and the space character.</p>
<p class="last">The reason the support for the <code>system</code> flag was removed is due to
the security implications with shell expansions when executing scripts
via the <code>system()</code> call.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--setcon <var>context</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Apply SELinux <tt class="docutils literal">context</tt> after initialization. This essentially
provides the ability to restrict OpenVPN's rights to only network I/O
operations, thanks to SELinux. This goes further than <tt class="docutils literal"><span class="pre">--user</span></tt> and
<tt class="docutils literal"><span class="pre">--chroot</span></tt> in that those two, while being great security features,
unfortunately do not protect against privilege escalation by
exploitation of a vulnerable system call. You can of course combine all
three, but please note that since setcon requires access to /proc you
will have to provide it inside the chroot directory (e.g. with mount
--bind).</p>
<p>Since the setcon operation is delayed until after initialization,
OpenVPN can be restricted to just network-related system calls, whereas
by applying the context before startup (such as the OpenVPN one provided
in the SELinux Reference Policies) you will have to allow many things
required only during initialization.</p>
<p class="last">Like with chroot, complications can result when scripts or restarts are
executed after the setcon operation, which is why you should really
consider using the <tt class="docutils literal"><span class="pre">--persist-key</span></tt> and <tt class="docutils literal"><span class="pre">--persist-tun</span></tt> options.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--status <var>args</var></span></kbd></td>
<td><p class="first">Write operational status to <tt class="docutils literal">file</tt> every <tt class="docutils literal">n</tt> seconds. <tt class="docutils literal">n</tt> defaults
to <code>60</code> if not specified.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
status file
status file n
</pre>
<p>Status can also be written to the syslog by sending a <code>SIGUSR2</code>
signal.</p>
<p>With multi-client capability enabled on a server, the status file
includes a list of clients and a routing table. The output format can be
controlled by the <tt class="docutils literal"><span class="pre">--status-version</span></tt> option in that case.</p>
<p class="last">For clients or instances running in point-to-point mode, it will contain
the traffic statistics.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--status-version <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Set the status file format version number to <tt class="docutils literal">n</tt>.</p>
<p>This only affects the status file on servers with multi-client
capability enabled. Valid status version values:</p>
<dl class="last docutils">
<dt><code>1</code></dt>
<dd>Traditional format (default). The client list contains the
following fields comma-separated: Common Name, Real Address, Bytes
Received, Bytes Sent, Connected Since.</dd>
<dt><code>2</code></dt>
<dd>A more reliable format for external processing. Compared to
version <code>1</code>, the client list contains some additional fields:
Virtual Address, Virtual IPv6 Address, Username, Client ID, Peer ID,
Data Channel Cipher. Future versions may extend the number of fields.</dd>
<dt><code>3</code></dt>
<dd>Identical to <code>2</code>, but fields are tab-separated.</dd>
</dl>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--test-crypto</span></kbd></td>
<td><p class="first">Do a self-test of OpenVPN's crypto options by encrypting and decrypting
test packets using the data channel encryption options specified above.
This option does not require a peer to function, and therefore can be
specified without <tt class="docutils literal"><span class="pre">--dev</span></tt> or <tt class="docutils literal"><span class="pre">--remote</span></tt>.</p>
<p>The typical usage of <tt class="docutils literal"><span class="pre">--test-crypto</span></tt> would be something like this:</p>
<pre class="literal-block">
openvpn --test-crypto --secret key
</pre>
<p>or</p>
<pre class="literal-block">
openvpn --test-crypto --secret key --verb 9
</pre>
<p class="last">This option is very useful to test OpenVPN after it has been ported to a
new platform, or to isolate problems in the compiler, OpenSSL crypto
library, or OpenVPN's crypto code. Since it is a self-test mode,
problems with encryption and authentication can be debugged
independently of network and tunnel issues.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--tmp-dir <var>dir</var></span></kbd></td>
<td><p class="first">Specify a directory <tt class="docutils literal">dir</tt> for temporary files. This directory will be
used by openvpn processes and script to communicate temporary data with
openvpn main process. Note that the directory must be writable by the
OpenVPN process after it has dropped it's root privileges.</p>
<p>This directory will be used by in the following cases:</p>
<ul class="last simple">
<li><tt class="docutils literal"><span class="pre">--client-connect</span></tt> scripts and <code>OPENVPN_PLUGIN_CLIENT_CONNECT</code>
plug-in hook to dynamically generate client-specific configuration
<code>client_connect_config_file</code> and return success/failure via
<code>client_connect_deferred_file</code> when using deferred client connect
method</li>
<li><code>OPENVPN_PLUGIN_AUTH_USER_PASS_VERIFY</code> plug-in hooks returns
success/failure via <code>auth_control_file</code> when using deferred auth
method and pending authentication via <code>pending_auth_file</code>.</li>
</ul>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--use-prediction-resistance</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Enable prediction resistance on mbed TLS's RNG.</p>
<p>Enabling prediction resistance causes the RNG to reseed in each call for
random. Reseeding this often can quickly deplete the kernel entropy
pool.</p>
<p class="last">If you need this option, please consider running a daemon that adds
entropy to the kernel pool.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--user <var>user</var></span></kbd></td>
<td><p class="first">Change the user ID of the OpenVPN process to <tt class="docutils literal">user</tt> after
initialization, dropping privileges in the process. This option is
useful to protect the system in the event that some hostile party was
able to gain control of an OpenVPN session. Though OpenVPN's security
features make this unlikely, it is provided as a second line of defense.</p>
<p>By setting <tt class="docutils literal">user</tt> to an unprivileged user dedicated to run openvpn,
the hostile party would be limited in what damage they could cause. Of
course once you take away privileges, you cannot return them to an
OpenVPN session. This means, for example, that if you want to reset an
OpenVPN daemon with a <code>SIGUSR1</code> signal (for example in response to
a DHCP reset), you should make use of one or more of the <tt class="docutils literal"><span class="pre">--persist</span></tt>
options to ensure that OpenVPN doesn't need to execute any privileged
operations in order to restart (such as re-reading key files or running
<tt class="docutils literal">ifconfig</tt> on the TUN device).</p>
<p class="last">NOTE: Previous versions of openvpn used <code>nobody</code> as the example
unpriviledged user. It is not recommended to actually use that user
since it is usually used by other system services already. Always
create a dedicated user for openvpn.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--writepid <var>file</var></span></kbd></td>
</tr>
<tr><td> </td><td>Write OpenVPN's main process ID to <tt class="docutils literal">file</tt>.</td></tr>
</tbody>
</table>
</div>
<div class="section" id="log-options">
<h2>Log options</h2>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group">
<kbd><span class="option">--echo <var>parms</var></span></kbd></td>
<td><p class="first">Echo <tt class="docutils literal">parms</tt> to log output.</p>
<p class="last">Designed to be used to send messages to a controlling application which
is receiving the OpenVPN log output.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--errors-to-stderr</span></kbd></td>
</tr>
<tr><td> </td><td>Output errors to stderr instead of stdout unless log output is
redirected by one of the <tt class="docutils literal"><span class="pre">--log</span></tt> options.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--log <var>file</var></span></kbd></td>
<td><p class="first">Output logging messages to <tt class="docutils literal">file</tt>, including output to stdout/stderr
which is generated by called scripts. If <tt class="docutils literal">file</tt> already exists it will
be truncated. This option takes effect immediately when it is parsed in
the command line and will supersede syslog output if <tt class="docutils literal"><span class="pre">--daemon</span></tt>
is also specified. This option is persistent over the entire
course of an OpenVPN instantiation and will not be reset by
<code>SIGHUP</code>, <code>SIGUSR1</code>, or <tt class="docutils literal"><span class="pre">--ping-restart</span></tt>.</p>
<p class="last">Note that on Windows, when OpenVPN is started as a service, logging
occurs by default without the need to specify this option.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--log-append <var>file</var></span></kbd></td>
</tr>
<tr><td> </td><td>Append logging messages to <tt class="docutils literal">file</tt>. If <tt class="docutils literal">file</tt> does not exist, it will
be created. This option behaves exactly like <tt class="docutils literal"><span class="pre">--log</span></tt> except that it
appends to rather than truncating the log file.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--machine-readable-output</span></kbd></td>
</tr>
<tr><td> </td><td>Always write timestamps and message flags to log messages, even when
they otherwise would not be prefixed. In particular, this applies to log
messages sent to stdout.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--mute <var>n</var></span></kbd></td>
<td>Log at most <tt class="docutils literal">n</tt> consecutive messages in the same category. This is
useful to limit repetitive logging of similar message types.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--mute-replay-warnings</span></kbd></td>
</tr>
<tr><td> </td><td>Silence the output of replay warnings, which are a common false alarm on
WiFi networks. This option preserves the security of the replay
protection code without the verbosity associated with warnings about
duplicate packets.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--suppress-timestamps</span></kbd></td>
</tr>
<tr><td> </td><td>Avoid writing timestamps to log messages, even when they otherwise would
be prepended. In particular, this applies to log messages sent to
stdout.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--syslog <var>progname</var></span></kbd></td>
</tr>
<tr><td> </td><td>Direct log output to system logger, but do not become a daemon. See
<tt class="docutils literal"><span class="pre">--daemon</span></tt> directive above for description of <tt class="docutils literal">progname</tt> parameter.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--verb <var>n</var></span></kbd></td>
<td><p class="first">Set output verbosity to <tt class="docutils literal">n</tt> (default <code>1</code>). Each level shows all
info from the previous levels. Level <code>3</code> is recommended if you want
a good summary of what's happening without being swamped by output.</p>
<dl class="last docutils">
<dt><code>0</code></dt>
<dd>No output except fatal errors.</dd>
<dt><code>1</code> to <code>4</code></dt>
<dd>Normal usage range.</dd>
<dt><code>5</code></dt>
<dd>Outputs <code>R</code> and <code>W</code> characters to the console for
each packet read and write, uppercase is used for TCP/UDP
packets and lowercase is used for TUN/TAP packets.</dd>
<dt><code>6</code> to <code>11</code></dt>
<dd>Debug info range (see <code>errlevel.h</code> in the source code for
additional information on debug levels).</dd>
</dl>
</td></tr>
</tbody>
</table>
</div>
<div class="section" id="protocol-options">
<h2>Protocol options</h2>
<p>Options in this section affect features available in the OpenVPN wire
protocol. Many of these options also define the encryption options
of the data channel in the OpenVPN wire protocol. These options must be
configured in a compatible way between both the local and remote side.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--allow-compression <var>mode</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">As described in the <tt class="docutils literal"><span class="pre">--compress</span></tt> option, compression is a potentially
dangerous option. This option allows controlling the behaviour of
OpenVPN when compression is used and allowed.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
allow-compression
allow-compression mode
</pre>
<p>The <tt class="docutils literal">mode</tt> argument can be one of the following values:</p>
<dl class="last docutils">
<dt><code>asym</code></dt>
<dd>OpenVPN will only <em>decompress downlink packets</em> but <em>not compress
uplink packets</em>. This also allows migrating to disable compression
when changing both server and client configurations to remove
compression at the same time is not a feasible option.</dd>
<dt><code>no</code> (default)</dt>
<dd>OpenVPN will refuse any compression. If data-channel offloading
is enabled, OpenVPN will additionally also refuse compression
framing (stub).</dd>
<dt><code>yes</code></dt>
<dd>OpenVPN will send and receive compressed packets.</dd>
</dl>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--auth <var>alg</var></span></kbd></td>
<td><p class="first">Authenticate data channel packets and (if enabled) <tt class="docutils literal"><span class="pre">tls-auth</span></tt> control
channel packets with HMAC using message digest algorithm <tt class="docutils literal">alg</tt>. (The
default is <tt class="docutils literal">SHA1</tt> ). HMAC is a commonly used message authentication
algorithm (MAC) that uses a data string, a secure hash algorithm and a
key to produce a digital signature.</p>
<p>The OpenVPN data channel protocol uses encrypt-then-mac (i.e. first
encrypt a packet then HMAC the resulting ciphertext), which prevents
padding oracle attacks.</p>
<p>If an AEAD cipher mode (e.g. GCM) is chosen then the specified <tt class="docutils literal"><span class="pre">--auth</span></tt>
algorithm is ignored for the data channel and the authentication method
of the AEAD cipher is used instead. Note that <tt class="docutils literal">alg</tt> still specifies
the digest used for <tt class="docutils literal"><span class="pre">tls-auth</span></tt>.</p>
<p>In static-key encryption mode, the HMAC key is included in the key file
generated by <tt class="docutils literal"><span class="pre">--genkey</span></tt>. In TLS mode, the HMAC key is dynamically
generated and shared between peers via the TLS control channel. If
OpenVPN receives a packet with a bad HMAC it will drop the packet. HMAC
usually adds 16 or 20 bytes per packet. Set <tt class="docutils literal">alg=none</tt> to disable
authentication.</p>
<p class="last">For more information on HMAC see
<a class="reference external" href="http://www.cs.ucsd.edu/users/mihir/papers/hmac.html">http://www.cs.ucsd.edu/users/mihir/papers/hmac.html</a></p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--cipher <var>alg</var></span></kbd></td>
<td><p class="first">This option should not be used any longer in TLS mode and still
exists for two reasons:</p>
<ul class="simple">
<li>compatibility with old configurations still carrying it
around;</li>
<li>allow users connecting to OpenVPN peers older than 2.6.0
to have <tt class="docutils literal"><span class="pre">--cipher</span></tt> configured the same way as the remote
counterpart. This can avoid MTU/frame size warnings.</li>
</ul>
<p>Before 2.4.0, this option was used to select the cipher to be
configured on the data channel, however, later versions usually
ignored this directive in favour of a negotiated cipher.
Starting with 2.6.0, this option is always ignored in TLS mode
when it comes to configuring the cipher and will only control the
cipher for <tt class="docutils literal"><span class="pre">--secret</span></tt> pre-shared-key mode (note: this mode is
deprecated and strictly not recommended).</p>
<p>If you wish to specify the cipher to use on the data channel,
please see <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> (for regular negotiation) and
<tt class="docutils literal"><span class="pre">--data-ciphers-fallback</span></tt> (for a fallback option when the
negotiation cannot take place because the other peer is old or
has negotiation disabled).</p>
<p>To see ciphers that are available with OpenVPN, use the
<tt class="docutils literal"><span class="pre">--show-ciphers</span></tt> option.</p>
<p class="last">Set <tt class="docutils literal">alg</tt> to <code>none</code> to disable encryption.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--compress <var>algorithm</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first"><strong>DEPRECATED</strong> Enable a compression algorithm. Compression is generally
not recommended. VPN tunnels which use compression are susceptible to
the VORALCE attack vector. See also the <code>migrate</code> parameter below.</p>
<p>The <tt class="docutils literal">algorithm</tt> parameter may be <code>lzo</code>, <code>lz4</code>,
<code>lz4-v2</code>, <code>stub</code>, <code>stub-v2</code>, <code>migrate</code> or empty.
LZO and LZ4 are different compression algorithms, with LZ4 generally
offering the best performance with least CPU usage.</p>
<p>The <code>lz4-v2</code> and <code>stub-v2</code> variants implement a better
framing that does not add overhead when packets cannot be compressed. All
other variants always add one extra framing byte compared to no
compression framing.</p>
<p>Especially <code>stub-v2</code> is essentially identical to no compression and
no compression framing as its header indicates IP version 5 in a tun setup
and can (ab)used to complete disable compression to clients. (See the
<code>migrate</code> option below)</p>
<p>If the <tt class="docutils literal">algorithm</tt> parameter is <code>stub</code>, <code>stub-v2</code> or empty,
compression will be turned off, but the packet framing for compression
will still be enabled, allowing a different setting to be pushed later.
Additionally, <code>stub</code> and <code>stub-v2</code> wil disable announcing
<tt class="docutils literal">lzo</tt> and <tt class="docutils literal">lz4</tt> compression support via <em>IV_</em> variables to the
server.</p>
<p>Note: the <code>stub</code> (or empty) option is NOT compatible with the older
option <tt class="docutils literal"><span class="pre">--comp-lzo</span> no</tt>.</p>
<p>Using <code>migrate</code> as compression algorithm enables a special migration mode.
It allows migration away from the <tt class="docutils literal"><span class="pre">--compress</span></tt>/<tt class="docutils literal"><span class="pre">--comp-lzo</span></tt> options to no compression.
This option sets the server to no compression mode and the server behaves identical to
a server without a compression option for all clients without a compression in their
config. However, if a client is detected that indicates that compression is used (via OCC),
the server will automatically add <tt class="docutils literal"><span class="pre">--push</span> compress <span class="pre">stub-v2</span></tt> to the client specific
configuration if supported by the client and otherwise switch to <tt class="docutils literal"><span class="pre">comp-lzo</span> no</tt>
and add <tt class="docutils literal"><span class="pre">--push</span> <span class="pre">comp-lzo</span></tt> to the client specific configuration.</p>
<p><strong>*Security Considerations*</strong></p>
<p class="last">Compression and encryption is a tricky combination. If an attacker knows
or is able to control (parts of) the plain-text of packets that contain
secrets, the attacker might be able to extract the secret if compression
is enabled. See e.g. the <em>CRIME</em> and <em>BREACH</em> attacks on TLS and
<em>VORACLE</em> on VPNs which also leverage to break encryption. If you are not
entirely sure that the above does not apply to your traffic, you are
advised to <em>not</em> enable compression.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--comp-lzo <var>mode</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first"><strong>DEPRECATED</strong> Enable LZO compression algorithm. Compression is
generally not recommended. VPN tunnels which uses compression are
suspectible to the VORALCE attack vector.</p>
<p>Use LZO compression -- may add up to 1 byte per packet for incompressible
data. <tt class="docutils literal">mode</tt> may be <code>yes</code>, <code>no</code>, or <code>adaptive</code>
(default).</p>
<p>In a server mode setup, it is possible to selectively turn compression
on or off for individual clients.</p>
<p>First, make sure the client-side config file enables selective
compression by having at least one <tt class="docutils literal"><span class="pre">--comp-lzo</span></tt> directive, such as
<tt class="docutils literal"><span class="pre">--comp-lzo</span> no</tt>. This will turn off compression by default, but allow
a future directive push from the server to dynamically change the
<code>on</code>/<code>off</code>/<code>adaptive</code> setting.</p>
<p>Next in a <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> file, specify the compression setting
for the client, for example:</p>
<pre class="literal-block">
comp-lzo yes
push "comp-lzo yes"
</pre>
<p class="last">The first line sets the <tt class="docutils literal"><span class="pre">comp-lzo</span></tt> setting for the server side of the
link, the second sets the client side.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--comp-noadapt</span></kbd></td>
<td><p class="first"><strong>DEPRECATED</strong> When used in conjunction with <tt class="docutils literal"><span class="pre">--comp-lzo</span></tt>, this option
will disable OpenVPN's adaptive compression algorithm. Normally, adaptive
compression is enabled with <tt class="docutils literal"><span class="pre">--comp-lzo</span></tt>.</p>
<p class="last">Adaptive compression tries to optimize the case where you have
compression enabled, but you are sending predominantly incompressible
(or pre-compressed) packets over the tunnel, such as an FTP or rsync
transfer of a large, compressed file. With adaptive compression, OpenVPN
will periodically sample the compression process to measure its
efficiency. If the data being sent over the tunnel is already
compressed, the compression efficiency will be very low, triggering
openvpn to disable compression for a period of time until the next
re-sample test.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--key-direction</span></kbd></td>
</tr>
<tr><td> </td><td>Alternative way of specifying the optional direction parameter for the
<tt class="docutils literal"><span class="pre">--tls-auth</span></tt> and <tt class="docutils literal"><span class="pre">--secret</span></tt> options. Useful when using inline files
(See section on inline files).</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--data-ciphers <var>cipher-list</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Restrict the allowed ciphers to be negotiated to the ciphers in
<tt class="docutils literal"><span class="pre">cipher-list</span></tt>. <tt class="docutils literal"><span class="pre">cipher-list</span></tt> is a colon-separated list of ciphers,
and defaults to <code>AES-256-GCM:AES-128-GCM:CHACHA20-POLY1305</code> when
Chacha20-Poly1305 is available and otherwise <code>AES-256-GCM:AES-128-GCM</code>.</p>
<p>For servers, the first cipher from <tt class="docutils literal"><span class="pre">cipher-list</span></tt> that is also
supported by the client will be pushed to clients that support cipher
negotiation.</p>
<p>For more details see the chapter on <a class="reference internal" href="#data-channel-cipher-negotiation">Data channel cipher negotiation</a>.
<em>Especially</em> if you need to support clients with OpenVPN versions older
than 2.4!</p>
<p>Starting with OpenVPN 2.6 a cipher can be prefixed with a <code>?</code> to mark
it as optional. This allows including ciphers in the list that may not be
available on all platforms.
E.g. <code>AES-256-GCM:AES-128-GCM:?CHACHA20-POLY1305</code> would only enable
Chacha20-Poly1305 if the underlying SSL library (and its configuration)
supports it.</p>
<p>Cipher negotiation is enabled in client-server mode only. I.e. if
<tt class="docutils literal"><span class="pre">--mode</span></tt> is set to <cite>server</cite> (server-side, implied by setting
<tt class="docutils literal"><span class="pre">--server</span></tt> ), or if <tt class="docutils literal"><span class="pre">--pull</span></tt> is specified (client-side, implied by
setting <tt class="docutils literal"><span class="pre">--client</span></tt>).</p>
<p>If no common cipher is found during cipher negotiation, the connection
is terminated. To support old clients/old servers that do not provide any
cipher negotiation support see <tt class="docutils literal"><span class="pre">--data-ciphers-fallback</span></tt>.</p>
<p>If <tt class="docutils literal"><span class="pre">--compat-mode</span></tt> is set to a version older than 2.5.0 the cipher
specified by <tt class="docutils literal"><span class="pre">--cipher</span></tt> will be appended to <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> if
not already present.</p>
<p>This list is restricted to be 127 chars long after conversion to OpenVPN
ciphers.</p>
<p class="last">This option was called <tt class="docutils literal"><span class="pre">--ncp-ciphers</span></tt> in OpenVPN 2.4 but has been renamed
to <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> in OpenVPN 2.5 to more accurately reflect its meaning.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--data-ciphers-fallback <var>alg</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Configure a cipher that is used to fall back to if we could not determine
which cipher the peer is willing to use.</p>
<p class="last">This option should only be needed to
connect to peers that are running OpenVPN 2.3 or older versions, and
have been configured with <tt class="docutils literal"><span class="pre">--enable-small</span></tt>
(typically used on routers or other embedded devices).</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--secret <var>args</var></span></kbd></td>
<td><p class="first"><strong>DEPRECATED</strong> Enable Static Key encryption mode (non-TLS). Use pre-shared secret
<tt class="docutils literal">file</tt> which was generated with <tt class="docutils literal"><span class="pre">--genkey</span></tt>.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
secret file
secret file direction
</pre>
<p>The optional <tt class="docutils literal">direction</tt> parameter enables the use of 4 distinct keys
(HMAC-send, cipher-encrypt, HMAC-receive, cipher-decrypt), so that each
data flow direction has a different set of HMAC and cipher keys. This
has a number of desirable security properties including eliminating
certain kinds of DoS and message replay attacks.</p>
<p>When the <tt class="docutils literal">direction</tt> parameter is omitted, 2 keys are used
bidirectionally, one for HMAC and the other for encryption/decryption.</p>
<p>The <tt class="docutils literal">direction</tt> parameter should always be complementary on either
side of the connection, i.e. one side should use <code>0</code> and the other
should use <code>1</code>, or both sides should omit it altogether.</p>
<p>The <tt class="docutils literal">direction</tt> parameter requires that <tt class="docutils literal">file</tt> contains a 2048 bit
key. While pre-1.5 versions of OpenVPN generate 1024 bit key files, any
version of OpenVPN which supports the <tt class="docutils literal">direction</tt> parameter, will also
support 2048 bit key file generation using the <tt class="docutils literal"><span class="pre">--genkey</span></tt> option.</p>
<p>Static key encryption mode has certain advantages, the primary being
ease of configuration.</p>
<p>There are no certificates or certificate authorities or complicated
negotiation handshakes and protocols. The only requirement is that you
have a pre-existing secure channel with your peer (such as <tt class="docutils literal">ssh</tt>) to
initially copy the key. This requirement, along with the fact that your
key never changes unless you manually generate a new one, makes it
somewhat less secure than TLS mode (see below). If an attacker manages
to steal your key, everything that was ever encrypted with it is
compromised. Contrast that to the perfect forward secrecy features of
TLS mode (using Diffie Hellman key exchange), where even if an attacker
was able to steal your private key, he would gain no information to help
him decrypt past sessions.</p>
<p class="last">Another advantageous aspect of Static Key encryption mode is that it is
a handshake-free protocol without any distinguishing signature or
feature (such as a header or protocol handshake sequence) that would
mark the ciphertext packets as being generated by OpenVPN. Anyone
eavesdropping on the wire would see nothing but random-looking data.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tran-window <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td>Transition window -- our old key can live this many seconds after a new
a key renegotiation begins (default <code>3600</code> seconds). This feature
allows for a graceful transition from old to new key, and removes the key
renegotiation sequence from the critical path of tunnel data forwarding.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--force-tls-key-material-export</span></kbd></td>
</tr>
<tr><td> </td><td>This option is only available in --mode server and forces to use
Keying Material Exporters (RFC 5705) for clients. This can be used to
simulate an environment where the cryptographic library does not support
the older method to generate data channel keys anymore. This option is
intended to be a test option and might be removed in a future OpenVPN
version without notice.</td></tr>
</tbody>
</table>
</div>
<div class="section" id="client-options">
<h2>Client Options</h2>
<p>The client options are used when connecting to an OpenVPN server configured
to use <tt class="docutils literal"><span class="pre">--server</span></tt>, <tt class="docutils literal"><span class="pre">--server-bridge</span></tt>, or <tt class="docutils literal"><span class="pre">--mode</span> server</tt> in its
configuration.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--allow-pull-fqdn</span></kbd></td>
</tr>
<tr><td> </td><td>Allow client to pull DNS names from server (rather than being limited to
IP address) for <tt class="docutils literal"><span class="pre">--ifconfig</span></tt>, <tt class="docutils literal"><span class="pre">--route</span></tt>, and <tt class="docutils literal"><span class="pre">--route-gateway</span></tt>.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--allow-recursive-routing</span></kbd></td>
</tr>
<tr><td> </td><td>When this option is set, OpenVPN will not drop incoming tun packets with
same destination as host.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--auth-token <var>token</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">This is not an option to be used directly in any configuration files,
but rather push this option from a <tt class="docutils literal"><span class="pre">--client-connect</span></tt> script or a
<tt class="docutils literal"><span class="pre">--plugin</span></tt> which hooks into the <code>OPENVPN_PLUGIN_CLIENT_CONNECT</code>
or <code>OPENVPN_PLUGIN_CLIENT_CONNECT_V2</code> calls. This option provides a
possibility to replace the clients password with an authentication token
during the lifetime of the OpenVPN client.</p>
<p>Whenever the connection is renegotiated and the
<tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> script or <tt class="docutils literal"><span class="pre">--plugin</span></tt> making use of the
<code>OPENVPN_PLUGIN_AUTH_USER_PASS_VERIFY</code> hook is triggered, it will
pass over this token as the password instead of the password the user
provided. The authentication token can only be reset by a full reconnect
where the server can push new options to the client. The password the
user entered is never preserved once an authentication token has been
set. If the OpenVPN server side rejects the authentication token then
the client will receive an <code>AUTH_FAILED</code> and disconnect.</p>
<p>The purpose of this is to enable two factor authentication methods, such
as HOTP or TOTP, to be used without needing to retrieve a new OTP code
each time the connection is renegotiated. Another use case is to cache
authentication data on the client without needing to have the users
password cached in memory during the life time of the session.</p>
<p>To make use of this feature, the <tt class="docutils literal"><span class="pre">--client-connect</span></tt> script or
<tt class="docutils literal"><span class="pre">--plugin</span></tt> needs to put</p>
<pre class="literal-block">
push "auth-token UNIQUE_TOKEN_VALUE"
</pre>
<p>into the file/buffer for dynamic configuration data. This will then make
the OpenVPN server to push this value to the client, which replaces the
local password with the <tt class="docutils literal">UNIQUE_TOKEN_VALUE</tt>.</p>
<p class="last">Newer clients (2.4.7+) will fall back to the original password method
after a failed auth. Older clients will keep using the token value and
react according to <tt class="docutils literal"><span class="pre">--auth-retry</span></tt></p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--auth-token-user <var>base64username</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Companion option to <tt class="docutils literal"><span class="pre">--auth-token</span></tt>. This options allows one to override
the username used by the client when reauthenticating with the <tt class="docutils literal"><span class="pre">auth-token</span></tt>.
It also allows one to use <tt class="docutils literal"><span class="pre">--auth-token</span></tt> in setups that normally do not use
username and password.</p>
<p class="last">The username has to be base64 encoded.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--auth-user-pass</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Authenticate with server using username/password.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
auth-user-pass
auth-user-pass up
</pre>
<p>If <tt class="docutils literal">up</tt> is present, it must be a file containing username/password on 2
lines. If the password line is missing, OpenVPN will prompt for one.</p>
<p>If <tt class="docutils literal">up</tt> is omitted, username/password will be prompted from the
console.</p>
<p class="last">The server configuration must specify an <tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt>
script to verify the username/password provided by the client.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--auth-retry <var>type</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Controls how OpenVPN responds to username/password verification errors
such as the client-side response to an <code>AUTH_FAILED</code> message from
the server or verification failure of the private key password.</p>
<p>Normally used to prevent auth errors from being fatal on the client
side, and to permit username/password requeries in case of error.</p>
<p>An <code>AUTH_FAILED</code> message is generated by the server if the client
fails <tt class="docutils literal"><span class="pre">--auth-user-pass</span></tt> authentication, or if the server-side
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> script returns an error status when the client
tries to connect.</p>
<p><tt class="docutils literal">type</tt> can be one of:</p>
<dl class="docutils">
<dt><code>none</code></dt>
<dd>Client will exit with a fatal error (this is the default).</dd>
<dt><code>nointeract</code></dt>
<dd>Client will retry the connection without requerying
for an <tt class="docutils literal"><span class="pre">--auth-user-pass</span></tt> username/password. Use this option for
unattended clients.</dd>
<dt><code>interact</code></dt>
<dd>Client will requery for an <tt class="docutils literal"><span class="pre">--auth-user-pass</span></tt>
username/password and/or private key password before attempting a
reconnection.</dd>
</dl>
<p class="last">Note that while this option cannot be pushed, it can be controlled from
the management interface.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--client</span></kbd></td>
<td><p class="first">A helper directive designed to simplify the configuration of OpenVPN's
client mode. This directive is equivalent to:</p>
<pre class="last literal-block">
pull
tls-client
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--client-nat <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">This pushable client option sets up a stateless one-to-one NAT rule on
packet addresses (not ports), and is useful in cases where routes or
ifconfig settings pushed to the client would create an IP numbering
conflict.</p>
<p>Examples:</p>
<pre class="literal-block">
client-nat snat 192.168.0.0/255.255.0.0
client-nat dnat 10.64.0.0/255.255.0.0
</pre>
<p><tt class="docutils literal">network/netmask</tt> (for example <code>192.168.0.0/255.255.0.0</code>) defines
the local view of a resource from the client perspective, while
<tt class="docutils literal">alias/netmask</tt> (for example <code>10.64.0.0/255.255.0.0</code>) defines the
remote view from the server perspective.</p>
<p>Use <code>snat</code> (source NAT) for resources owned by the client and
<code>dnat</code> (destination NAT) for remote resources.</p>
<p class="last">Set <tt class="docutils literal"><span class="pre">--verb</span> 6</tt> for debugging info showing the transformation of
src/dest addresses in packets.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--connect-retry <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Wait <tt class="docutils literal">n</tt> seconds between connection attempts (default <code>1</code>).
Repeated reconnection attempts are slowed down after 5 retries per
remote by doubling the wait time after each unsuccessful attempt.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
connect retry n
connect retry n max
</pre>
<p class="last">If the optional argument <tt class="docutils literal">max</tt> is specified, the maximum wait time in
seconds gets capped at that value (default <code>300</code>).</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--connect-retry-max <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td><tt class="docutils literal">n</tt> specifies the number of times each <tt class="docutils literal"><span class="pre">--remote</span></tt> or
<tt class="docutils literal"><connection></tt> entry is tried. Specifying <tt class="docutils literal">n</tt> as <code>1</code> would try
each entry exactly once. A successful connection resets the counter.
(default <em>unlimited</em>).</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--connect-timeout <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td>See <tt class="docutils literal"><span class="pre">--server-poll-timeout</span></tt>.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--dns <var>args</var></span></kbd></td>
<td><p class="first">Client DNS configuration to be used with the connection.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
dns search-domains domain [domain ...]
dns server n address addr[:port] [addr[:port] ...]
dns server n resolve-domains domain [domain ...]
dns server n dnssec yes|optional|no
dns server n transport DoH|DoT|plain
dns server n sni server-name
</pre>
<p>The <tt class="docutils literal"><span class="pre">--dns</span> <span class="pre">search-domains</span></tt> directive takes one or more domain names
to be added as DNS domain suffixes. If it is repeated multiple times within
a configuration the domains are appended, thus e.g. domain names pushed by
a server will amend locally defined ones.</p>
<p>The <tt class="docutils literal"><span class="pre">--dns</span> server</tt> directive is used to configure DNS server <tt class="docutils literal">n</tt>.
The server id <tt class="docutils literal">n</tt> must be a value between -128 and 127. For pushed
DNS server options it must be between 0 and 127. The server id is used
to group options and also for ordering the list of configured DNS servers;
lower numbers come first. DNS servers being pushed to a client replace
already configured DNS servers with the same server id.</p>
<p>The <tt class="docutils literal">address</tt> option configures the IPv4 and / or IPv6 address(es) of
the DNS server. Up to eight addresses can be specified per DNS server.
Optionally a port can be appended after a colon. IPv6 addresses need to
be enclosed in brackets if a port is appended.</p>
<p>The <tt class="docutils literal"><span class="pre">resolve-domains</span></tt> option takes one or more DNS domains used to define
a split-dns or dns-routing setup, where only the given domains are resolved
by the server. Systems which do not support fine grained DNS domain
configuration will ignore this setting.</p>
<p>The <tt class="docutils literal">dnssec</tt> option is used to configure validation of DNSSEC records.
While the exact semantics may differ for resolvers on different systems,
<tt class="docutils literal">yes</tt> likely makes validation mandatory, <tt class="docutils literal">no</tt> disables it, and <tt class="docutils literal">optional</tt>
uses it opportunistically.</p>
<p>The <tt class="docutils literal">transport</tt> option enables DNS-over-HTTPS (<tt class="docutils literal">DoH</tt>) or DNS-over-TLS (<tt class="docutils literal">DoT</tt>)
for a DNS server. The <tt class="docutils literal">sni</tt> option can be used with them to specify the
<tt class="docutils literal"><span class="pre">server-name</span></tt> for TLS server name indication.</p>
<p>Each server has to have at least one address configured for a configuration
to be valid. All the other options can be omitted.</p>
<p>Note that not all options may be supported on all platforms. As soon support
for different systems is implemented, information will be added here how
unsupported options are treated.</p>
<p class="last">The <tt class="docutils literal"><span class="pre">--dns</span></tt> option will eventually obsolete the <tt class="docutils literal"><span class="pre">--dhcp-option</span></tt> directive.
Until then it will replace configuration at the places <tt class="docutils literal"><span class="pre">--dhcp-option</span></tt> puts it,
so that <tt class="docutils literal"><span class="pre">--dns</span></tt> overrides <tt class="docutils literal"><span class="pre">--dhcp-option</span></tt>. Thus, <tt class="docutils literal"><span class="pre">--dns</span></tt> can be used today
to migrate from <tt class="docutils literal"><span class="pre">--dhcp-option</span></tt>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--explicit-exit-notify <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">In UDP client mode or point-to-point mode, send server/peer an exit
notification if tunnel is restarted or OpenVPN process is exited. In
client mode, on exit/restart, this option will tell the server to
immediately close its client instance object rather than waiting for a
timeout.</p>
<p>If both server and client support sending this message using the control
channel, the message will be sent as control-channel message. Otherwise
the message is sent as data-channel message, which will be ignored by
data-channel offloaded peers.</p>
<p>The <strong>n</strong> parameter (default <code>1</code> if not present) controls the
maximum number of attempts that the client will try to resend the exit
notification message if messages are sent in data-channel mode.</p>
<p>In UDP server mode, send <code>RESTART</code> control channel command to
connected clients. The <tt class="docutils literal">n</tt> parameter (default <code>1</code> if not present)
controls client behavior. With <tt class="docutils literal">n</tt> = <code>1</code> client will attempt to
reconnect to the same server, with <tt class="docutils literal">n</tt> = <code>2</code> client will advance
to the next server.</p>
<p class="last">OpenVPN will not send any exit notifications unless this option is
enabled.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--inactive <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Causes OpenVPN to exit after <tt class="docutils literal">n</tt> seconds of inactivity on the TUN/TAP
device. The time length of inactivity is measured since the last
incoming or outgoing tunnel packet. The default value is 0 seconds,
which disables this feature.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
inactive n
inactive n bytes
</pre>
<p>If the optional <tt class="docutils literal">bytes</tt> parameter is included, exit if less than
<tt class="docutils literal">bytes</tt> of combined in/out traffic are produced on the tun/tap device
in <tt class="docutils literal">n</tt> seconds.</p>
<p class="last">In any case, OpenVPN's internal ping packets (which are just keepalives)
and TLS control packets are not considered "activity", nor are they
counted as traffic, as they are used internally by OpenVPN and are not
an indication of actual user activity.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--proto-force <var>p</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">When iterating through connection profiles, only consider profiles using
protocol <tt class="docutils literal">p</tt> (<code>tcp</code> | <code>udp</code>).</p>
<p>Note that this specifically only filters by the transport layer
protocol, i.e. UDP or TCP. This does not affect whether IPv4 or
IPv6 is used as IP protocol.</p>
<p class="last">For implementation reasons the option accepts the <code>4</code> and <code>6</code>
suffixes when specifying the protocol
(i.e. <code>udp4</code> / <code>udp6</code> / <code>tcp4</code> / <code>tcp6</code>).
However, these behave the same as without the suffix and should be avoided
to prevent confusion.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--pull</span></kbd></td>
<td><p class="first">This option must be used on a client which is connecting to a
multi-client server. It indicates to OpenVPN that it should accept
options pushed by the server, provided they are part of the legal set of
pushable options (note that the <tt class="docutils literal"><span class="pre">--pull</span></tt> option is implied by
<tt class="docutils literal"><span class="pre">--client</span></tt> ).</p>
<p class="last">In particular, <tt class="docutils literal"><span class="pre">--pull</span></tt> allows the server to push routes to the
client, so you should not use <tt class="docutils literal"><span class="pre">--pull</span></tt> or <tt class="docutils literal"><span class="pre">--client</span></tt> in situations
where you don't trust the server to have control over the client's
routing table.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--pull-filter <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Filter options on the client pushed by the server to the client.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
pull-filter accept text
pull-filter ignore text
pull-filter reject text
</pre>
<p>Filter options received from the server if the option starts with
<code>text</code>. The action flag <code>accept</code> allows the option,
<code>ignore</code> removes it and <code>reject</code> flags an error and triggers
a <code>SIGUSR1</code> restart. The filters may be specified multiple times,
and each filter is applied in the order it is specified. The filtering of
each option stops as soon as a match is found. Unmatched options are accepted
by default.</p>
<p>Prefix comparison is used to match <code>text</code> against the received option so
that</p>
<pre class="literal-block">
pull-filter ignore "route"
</pre>
<p>would remove all pushed options starting with <tt class="docutils literal">route</tt> which would
include, for example, <tt class="docutils literal"><span class="pre">route-gateway</span></tt>. Enclose <em>text</em> in quotes to
embed spaces.</p>
<pre class="literal-block">
pull-filter accept "route 192.168.1."
pull-filter ignore "route "
</pre>
<p>would remove all routes that do not start with <tt class="docutils literal">192.168.1</tt>.</p>
<p class="last"><em>Note</em> that <code>reject</code> may result in a repeated cycle of failure and
reconnect, unless multiple remotes are specified and connection to the
next remote succeeds. To silently ignore an option pushed by the server,
use <code>ignore</code>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--push-peer-info</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Push additional information about the client to server. The following
data is always pushed to the server:</p>
<dl class="docutils">
<dt><code>IV_VER=<version></code></dt>
<dd>The client OpenVPN version</dd>
<dt><code>IV_PLAT=[linux|solaris|openbsd|mac|netbsd|freebsd|win]</code></dt>
<dd>The client OS platform</dd>
<dt><code>IV_PROTO</code></dt>
<dd><p class="first">Details about protocol extensions that the peer supports. The
variable is a bitfield and the bits are defined as follows:</p>
<ul class="last simple">
<li>bit 0: Reserved, should always be zero</li>
<li>bit 1: The peer supports peer-id floating mechanism</li>
<li>bit 2: The client expects a push-reply and the server may
send this reply without waiting for a push-request first.</li>
<li>bit 3: The client is capable of doing key derivation using
RFC5705 key material exporter.</li>
<li>bit 4: The client is capable of accepting additional arguments
to the <tt class="docutils literal">AUTH_PENDING</tt> message.</li>
<li>bit 5: The client supports doing feature negotiation in P2P mode</li>
<li>bit 6: The client is capable of parsing and receiving the <tt class="docutils literal"><span class="pre">--dns</span></tt> pushed option</li>
<li>bit 7: The client is capable of sending exit notification via control channel using <tt class="docutils literal">EXIT</tt> message. Also, the client is accepting the protocol-flags pushed option for the EKM capability</li>
<li>bit 8: The client is capable of accepting <tt class="docutils literal">AUTH_FAILED,TEMP</tt> messages</li>
<li>bit 9: The client is capable of dynamic tls-crypt</li>
</ul>
</dd>
<dt><code>IV_NCP=2</code></dt>
<dd>Negotiable ciphers, client supports <tt class="docutils literal"><span class="pre">--cipher</span></tt> pushed by
the server, a value of 2 or greater indicates client supports
<em>AES-GCM-128</em> and <em>AES-GCM-256</em>. IV_NCP is <em>deprecated</em> in
favor of <tt class="docutils literal">IV_CIPHERS</tt>.</dd>
<dt><code>IV_CIPHERS=<data-ciphers></code></dt>
<dd>The client announces the list of supported ciphers configured with the
<tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> option to the server.</dd>
<dt><code>IV_MTU=<max_mtu></code></dt>
<dd>The client announces the support of pushable MTU and the maximum MTU
it is willing to accept.</dd>
<dt><code>IV_GUI_VER=<gui_id> <version></code></dt>
<dd>The UI version of a UI if one is running, for example
<code>de.blinkt.openvpn 0.5.47</code> for the Android app.
This may be set by the client UI/GUI using <tt class="docutils literal"><span class="pre">--setenv</span></tt>.</dd>
<dt><code>IV_SSO=[crtext,][openurl,][proxy_url]</code></dt>
<dd>Additional authentication methods supported by the client.
This may be set by the client UI/GUI using <tt class="docutils literal"><span class="pre">--setenv</span></tt>.</dd>
</dl>
<p>The following flags depend on which compression formats are compiled in
and whether compression is allowed by options. See <a class="reference internal" href="#protocol-options">Protocol options</a>
for more details.</p>
<blockquote>
<dl class="docutils">
<dt><code>IV_LZO=1</code></dt>
<dd>If client supports LZO compression.</dd>
<dt><code>IV_LZO_STUB=1</code></dt>
<dd>If client was built with LZO stub capability. This is only sent if
<tt class="docutils literal">IV_LZO=1</tt> is not sent. This means the client can talk to a server
configured with <tt class="docutils literal"><span class="pre">--comp-lzo</span> no</tt>.</dd>
<dt><code>IV_LZ4=1</code> and <code>IV_LZ4v2=1</code></dt>
<dd>If the client supports LZ4 compression.</dd>
<dt><code>IV_COMP_STUB=1</code> and <code>IV_COMP_STUBv2=1</code></dt>
<dd>If the client supports stub compression. This means the client can talk
to a server configured with <tt class="docutils literal"><span class="pre">--compress</span></tt>.</dd>
</dl>
</blockquote>
<p>When <tt class="docutils literal"><span class="pre">--push-peer-info</span></tt> is enabled the additional information consists
of the following data:</p>
<dl class="last docutils">
<dt><code>IV_HWADDR=<string></code></dt>
<dd>This is intended to be a unique and persistent ID of the client.
The string value can be any readable ASCII string up to 64 bytes.
OpenVPN 2.x and some other implementations use the MAC address of
the client's interface used to reach the default gateway. If this
string is generated by the client, it should be consistent and
preserved across independent sessions and preferably
re-installations and upgrades.</dd>
<dt><code>IV_SSL=<version string></code></dt>
<dd>The ssl library version used by the client, e.g.
<code>OpenSSL 1.0.2f 28 Jan 2016</code>.</dd>
<dt><code>IV_PLAT_VER=x.y</code></dt>
<dd>The version of the operating system, e.g. 6.1 for Windows 7.
This may be set by the client UI/GUI using <tt class="docutils literal"><span class="pre">--setenv</span></tt>.
On Windows systems it is automatically determined by openvpn
itself.</dd>
<dt><code>UV_<name>=<value></code></dt>
<dd>Client environment variables whose names start with
<code>UV_</code></dd>
</dl>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--remote <var>args</var></span></kbd></td>
<td><p class="first">Remote host name or IP address, port and protocol.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
remote host
remote host port
remote host port proto
</pre>
<p>The <tt class="docutils literal">port</tt> and <tt class="docutils literal">proto</tt> arguments are optional. The OpenVPN client
will try to connect to a server at <tt class="docutils literal">host:port</tt>. The <tt class="docutils literal">proto</tt> argument
indicates the protocol to use when connecting with the remote, and may be
<code>tcp</code> or <code>udp</code>. To enforce IPv4 or IPv6 connections add a
<code>4</code> or <code>6</code> suffix; like <code>udp4</code> / <code>udp6</code>
/ <code>tcp4</code> / <code>tcp6</code>.</p>
<p>On the client, multiple <tt class="docutils literal"><span class="pre">--remote</span></tt> options may be specified for
redundancy, each referring to a different OpenVPN server, in the order
specified by the list of <tt class="docutils literal"><span class="pre">--remote</span></tt> options. Specifying multiple
<tt class="docutils literal"><span class="pre">--remote</span></tt> options for this purpose is a special case of the more
general connection-profile feature. See the <tt class="docutils literal"><connection></tt>
documentation below.</p>
<p>The client will move on to the next host in the list, in the event of
connection failure. Note that at any given time, the OpenVPN client will
at most be connected to one server.</p>
<p>Examples:</p>
<pre class="literal-block">
remote server1.example.net
remote server1.example.net 1194
remote server2.example.net 1194 tcp
</pre>
<dl class="docutils">
<dt><em>Note:</em></dt>
<dd><p class="first">Since UDP is connectionless, connection failure is defined by
the <tt class="docutils literal"><span class="pre">--ping</span></tt> and <tt class="docutils literal"><span class="pre">--ping-restart</span></tt> options.</p>
<p class="last">Also, if you use multiple <tt class="docutils literal"><span class="pre">--remote</span></tt> options, AND you are dropping
root privileges on the client with <tt class="docutils literal"><span class="pre">--user</span></tt> and/or <tt class="docutils literal"><span class="pre">--group</span></tt> AND
the client is running a non-Windows OS, if the client needs to switch
to a different server, and that server pushes back different TUN/TAP
or route settings, the client may lack the necessary privileges to
close and reopen the TUN/TAP interface. This could cause the client
to exit with a fatal error.</p>
</dd>
</dl>
<p>If <tt class="docutils literal"><span class="pre">--remote</span></tt> is unspecified, OpenVPN will listen for packets from any
IP address, but will not act on those packets unless they pass all
authentication tests. This requirement for authentication is binding on
all potential peers, even those from known and supposedly trusted IP
addresses (it is very easy to forge a source IP address on a UDP
packet).</p>
<p>When used in TCP mode, <tt class="docutils literal"><span class="pre">--remote</span></tt> will act as a filter, rejecting
connections from any host which does not match <tt class="docutils literal">host</tt>.</p>
<p class="last">If <tt class="docutils literal">host</tt> is a DNS name which resolves to multiple IP addresses,
OpenVPN will try them in the order that the system getaddrinfo()
presents them, so priorization and DNS randomization is done by the
system library. Unless an IP version is forced by the protocol
specification (4/6 suffix), OpenVPN will try both IPv4 and IPv6
addresses, in the order getaddrinfo() returns them.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--remote-random</span></kbd></td>
</tr>
<tr><td> </td><td>When multiple <tt class="docutils literal"><span class="pre">--remote</span></tt> address/ports are specified, or if connection
profiles are being used, initially randomize the order of the list as a
kind of basic load-balancing measure.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--remote-random-hostname</span></kbd></td>
</tr>
<tr><td> </td><td>Prepend a random string (6 bytes, 12 hex characters) to hostname to
prevent DNS caching. For example, "foo.bar.gov" would be modified to
"<random-chars>.foo.bar.gov".</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--resolv-retry <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">If hostname resolve fails for <tt class="docutils literal"><span class="pre">--remote</span></tt>, retry resolve for <tt class="docutils literal">n</tt>
seconds before failing.</p>
<p>Set <tt class="docutils literal">n</tt> to <code>infinite</code> to retry indefinitely.</p>
<p class="last">By default, <tt class="docutils literal"><span class="pre">--resolv-retry</span> infinite</tt> is enabled. You can disable by
setting n=0.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--single-session</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">After initially connecting to a remote peer, disallow any new
connections. Using this option means that a remote peer cannot connect,
disconnect, and then reconnect.</p>
<p>If the daemon is reset by a signal or <tt class="docutils literal"><span class="pre">--ping-restart</span></tt>, it will allow
one new connection.</p>
<p class="last"><tt class="docutils literal"><span class="pre">--single-session</span></tt> can be used with <tt class="docutils literal"><span class="pre">--ping-exit</span></tt> or <tt class="docutils literal"><span class="pre">--inactive</span></tt>
to create a single dynamic session that will exit when finished.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--server-poll-timeout <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td>When connecting to a remote server do not wait for more than <tt class="docutils literal">n</tt>
seconds for a response before trying the next server. The default value
is <code>120</code>. This timeout includes proxy and TCP connect timeouts.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--static-challenge <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Enable static challenge/response protocol</p>
<p>Valid syntax:</p>
<pre class="literal-block">
static-challenge text echo
</pre>
<p>The <tt class="docutils literal">text</tt> challenge text is presented to the user which describes what
information is requested. The <tt class="docutils literal">echo</tt> flag indicates if the user's
input should be echoed on the screen. Valid <tt class="docutils literal">echo</tt> values are
<code>0</code> or <code>1</code>.</p>
<p class="last">See management-notes.txt in the OpenVPN distribution for a description of
the OpenVPN challenge/response protocol.</p>
</td></tr>
</tbody>
</table>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--http-proxy <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Connect to remote host through an HTTP proxy. This requires at least an
address <tt class="docutils literal">server</tt> and <tt class="docutils literal">port</tt> argument. If HTTP Proxy-Authenticate
is required, a file name to an <tt class="docutils literal">authfile</tt> file containing a username
and password on 2 lines can be given, or <code>stdin</code> to prompt from
console. Its content can also be specified in the config file with the
<tt class="docutils literal"><span class="pre">--http-proxy-user-pass</span></tt> option (See <a class="reference internal" href="#inline-file-support">INLINE FILE SUPPORT</a>).</p>
<p>The last optional argument is an <tt class="docutils literal"><span class="pre">auth-method</span></tt> which should be one
of <code>none</code>, <code>basic</code>, or <code>ntlm</code>.</p>
<p>HTTP Digest authentication is supported as well, but only via the
<code>auto</code> or <code>auto-nct</code> flags (below). This must replace
the <tt class="docutils literal">authfile</tt> argument.</p>
<p>The <code>auto</code> flag causes OpenVPN to automatically determine the
<tt class="docutils literal"><span class="pre">auth-method</span></tt> and query stdin or the management interface for
username/password credentials, if required. This flag exists on OpenVPN
2.1 or higher.</p>
<p>The <tt class="docutils literal"><span class="pre">auto-nct</span></tt> flag (no clear-text auth) instructs OpenVPN to
automatically determine the authentication method, but to reject weak
authentication protocols such as HTTP Basic Authentication.</p>
<p>Examples:</p>
<pre class="last literal-block">
# no authentication
http-proxy proxy.example.net 3128
# basic authentication, load credentials from file
http-proxy proxy.example.net 3128 authfile.txt
# basic authentication, ask user for credentials
http-proxy proxy.example.net 3128 stdin
# NTLM authentication, load credentials from file
http-proxy proxy.example.net 3128 authfile.txt ntlm2
# determine which authentication is required, ask user for credentials
http-proxy proxy.example.net 3128 auto
# determine which authentication is required, but reject basic
http-proxy proxy.example.net 3128 auto-nct
# determine which authentication is required, but set credentials
http-proxy proxy.example.net 3128 auto
http-proxy-user-pass authfile.txt
# basic authentication, specify credentials inline
http-proxy proxy.example.net 3128 "" basic
<http-proxy-user-pass>
username
password
</http-proxy-user-pass>
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--http-proxy-user-pass <var>userpass</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Overwrite the username/password information for <tt class="docutils literal"><span class="pre">--http-proxy</span></tt>. If specified
as an inline option (see <a class="reference internal" href="#inline-file-support">INLINE FILE SUPPORT</a>), it will be interpreted as
username/password separated by a newline. When specified on the command line
it is interpreted as a filename same as the third argument to <tt class="docutils literal"><span class="pre">--http-proxy</span></tt>.</p>
<p>Example:</p>
<pre class="last literal-block">
<http-proxy-user-pass>
username
password
</http-proxy-user-pass>
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--http-proxy-option <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Set extended HTTP proxy options. Requires an option <tt class="docutils literal">type</tt> as argument
and an optional <tt class="docutils literal">parameter</tt> to the type. Repeat to set multiple
options.</p>
<dl class="docutils">
<dt><code>VERSION</code> <tt class="docutils literal">version</tt></dt>
<dd>Set HTTP version number to <tt class="docutils literal">version</tt> (default <code>1.0</code>).</dd>
<dt><code>AGENT</code> <tt class="docutils literal"><span class="pre">user-agent</span></tt></dt>
<dd>Set HTTP "User-Agent" string to <tt class="docutils literal"><span class="pre">user-agent</span></tt>.</dd>
<dt><code>CUSTOM-HEADER</code> <tt class="docutils literal">name</tt> <tt class="docutils literal">content</tt></dt>
<dd>Adds the custom Header with <tt class="docutils literal">name</tt> as name and <tt class="docutils literal">content</tt> as
the content of the custom HTTP header.</dd>
</dl>
<p>Examples:</p>
<pre class="last literal-block">
http-proxy-option VERSION 1.1
http-proxy-option AGENT OpenVPN/2.4
http-proxy-option X-Proxy-Flag some-flags
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--socks-proxy <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td>Connect to remote host through a Socks5 proxy. A required <tt class="docutils literal">server</tt>
argument is needed. Optionally a <tt class="docutils literal">port</tt> (default <code>1080</code>) and
<tt class="docutils literal">authfile</tt> can be given. The <tt class="docutils literal">authfile</tt> is a file containing a
username and password on 2 lines, or <code>stdin</code> can be used to
prompt from console.</td></tr>
</tbody>
</table>
</div>
<div class="section" id="server-options">
<h2>Server Options</h2>
<p>Starting with OpenVPN 2.0, a multi-client TCP/UDP server mode is
supported, and can be enabled with the <tt class="docutils literal"><span class="pre">--mode</span> server</tt> option. In
server mode, OpenVPN will listen on a single port for incoming client
connections. All client connections will be routed through a single tun
or tap interface. This mode is designed for scalability and should be
able to support hundreds or even thousands of clients on sufficiently
fast hardware. SSL/TLS authentication must be used in this mode.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--auth-gen-token <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Returns an authentication token to successfully authenticated clients.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
auth-gen-token [lifetime] [renewal-time] [external-auth]
</pre>
<p>After successful user/password authentication, the OpenVPN server will
with this option generate a temporary authentication token and push that
to the client. On the following renegotiations, the OpenVPN client will pass
this token instead of the users password. On the server side the server
will do the token authentication internally and it will NOT do any
additional authentications against configured external user/password
authentication mechanisms.</p>
<p>The tokens implemented by this mechanism include an initial timestamp and
a renew timestamp and are secured by HMAC.</p>
<p>The <tt class="docutils literal">lifetime</tt> argument defines how long the generated token is valid.
The lifetime is defined in seconds. If lifetime is not set or it is set
to <code>0</code>, the token will never expire.</p>
<p>If <tt class="docutils literal"><span class="pre">renewal-time</span></tt> is not set it defaults to <tt class="docutils literal"><span class="pre">reneg-sec</span></tt>.</p>
<p>The token will expire either after the configured <tt class="docutils literal">lifetime</tt> of the
token is reached or after not being renewed for more than 2 *
<tt class="docutils literal"><span class="pre">renewal-time</span></tt> seconds. Clients will be sent renewed tokens on every TLS
renegotiation. If <tt class="docutils literal"><span class="pre">renewal-time</span></tt> is lower than <tt class="docutils literal"><span class="pre">reneg-sec</span></tt> the server
will push an updated temporary authentication token every <tt class="docutils literal"><span class="pre">reneweal-time</span></tt>
seconds. This is done to invalidate a token if a client is disconnected for a
sufficiently long time, while at the same time permitting much longer token
lifetimes for active clients.</p>
<p>This feature is useful for environments which are configured to use One
Time Passwords (OTP) as part of the user/password authentications and
that authentication mechanism does not implement any auth-token support.</p>
<p>When the <code>external-auth</code> keyword is present the normal
authentication method will always be called even if auth-token succeeds.
Normally other authentications method are skipped if auth-token
verification succeeds or fails.</p>
<p>This option postpones this decision to the external authentication
methods and checks the validity of the account and do other checks.</p>
<p>In this mode the environment will have a <tt class="docutils literal">session_id</tt> variable that
holds the session id from auth-gen-token. Also an environment variable
<tt class="docutils literal">session_state</tt> is present. This variable indicates whether the
auth-token has succeeded or not. It can have the following values:</p>
<dl class="docutils">
<dt><code>Initial</code></dt>
<dd>No token from client.</dd>
<dt><code>Authenticated</code></dt>
<dd>Token is valid and not expired.</dd>
<dt><code>Expired</code></dt>
<dd>Token is valid but has expired.</dd>
<dt><code>Invalid</code></dt>
<dd>Token is invalid (failed HMAC or wrong length)</dd>
<dt><code>AuthenticatedEmptyUser</code> / <code>ExpiredEmptyUser</code></dt>
<dd>The token is not valid with the username sent from the client but
would be valid (or expired) if we assume an empty username was
used instead. These two cases are a workaround for behaviour in
OpenVPN 3. If this workaround is not needed these two cases should
be handled in the same way as <code>Invalid</code>.</dd>
</dl>
<p class="last"><strong>Warning:</strong> Use this feature only if you want your authentication
method called on every verification. Since the external authentication
is called it needs to also indicate a success or failure of the
authentication. It is strongly recommended to return an authentication
failure in the case of the Invalid/Expired auth-token with the
external-auth option unless the client could authenticate in another
acceptable way (e.g. client certificate), otherwise returning success
will lead to authentication bypass (as does returning success on a wrong
password from a script).</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--auth-gen-token-secret <var>file</var></span></kbd></td>
</tr>
<tr><td> </td><td>Specifies a file that holds a secret for the HMAC used in
<tt class="docutils literal"><span class="pre">--auth-gen-token</span></tt> If <tt class="docutils literal">file</tt> is not present OpenVPN will generate a
random secret on startup. This file should be used if auth-token should
validate after restarting a server or if client should be able to roam
between multiple OpenVPN servers with their auth-token.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--auth-user-pass-optional</span></kbd></td>
</tr>
<tr><td> </td><td>Allow connections by clients that do not specify a username/password.
Normally, when <tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> or
<tt class="docutils literal"><span class="pre">--management-client-auth</span></tt> are specified (or an authentication plugin
module), the OpenVPN server daemon will require connecting clients to
specify a username and password. This option makes the submission of a
username/password by clients optional, passing the responsibility to the
user-defined authentication module/script to accept or deny the client
based on other factors (such as the setting of X509 certificate fields).
When this option is used, and a connecting client does not submit a
username/password, the user-defined authentication module/script will
see the username and password as being set to empty strings (""). The
authentication module/script MUST have logic to detect this condition
and respond accordingly.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ccd-exclusive</span></kbd></td>
</tr>
<tr><td> </td><td>Require, as a condition of authentication, that a connecting client has
a <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> file.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--client-config-dir <var>dir</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Specify a directory <tt class="docutils literal">dir</tt> for custom client config files. After a
connecting client has been authenticated, OpenVPN will look in this
directory for a file having the same name as the client's X509 common
name. If a matching file exists, it will be opened and parsed for
client-specific configuration options. If no matching file is found,
OpenVPN will instead try to open and parse a default file called
"DEFAULT", which may be provided but is not required. Note that the
configuration files must be readable by the OpenVPN process after it has
dropped it's root privileges.</p>
<p>This file can specify a fixed IP address for a given client using
<tt class="docutils literal"><span class="pre">--ifconfig-push</span></tt>, as well as fixed subnets owned by the client using
<tt class="docutils literal"><span class="pre">--iroute</span></tt>.</p>
<p>One of the useful properties of this option is that it allows client
configuration files to be conveniently created, edited, or removed while
the server is live, without needing to restart the server.</p>
<p class="last">The following options are legal in a client-specific context: <tt class="docutils literal"><span class="pre">--push</span></tt>,
<tt class="docutils literal"><span class="pre">--push-reset</span></tt>, <tt class="docutils literal"><span class="pre">--push-remove</span></tt>, <tt class="docutils literal"><span class="pre">--iroute</span></tt>, <tt class="docutils literal"><span class="pre">--ifconfig-push</span></tt>,
<tt class="docutils literal"><span class="pre">--vlan-pvid</span></tt> and <tt class="docutils literal"><span class="pre">--config</span></tt>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--client-to-client</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Because the OpenVPN server mode handles multiple clients through a
single tun or tap interface, it is effectively a router. The
<tt class="docutils literal"><span class="pre">--client-to-client</span></tt> flag tells OpenVPN to internally route
client-to-client traffic rather than pushing all client-originating
traffic to the TUN/TAP interface.</p>
<p>When this option is used, each client will "see" the other clients which
are currently connected. Otherwise, each client will only see the
server. Don't use this option if you want to firewall tunnel traffic
using custom, per-client rules.</p>
<p class="last">Please note that when using data channel offload this option has no
effect. Packets are always sent to the tunnel interface and then
routed based on the system routing table.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--disable</span></kbd></td>
<td><p class="first">Disable a particular client (based on the common name) from connecting.
Don't use this option to disable a client due to key or password
compromise. Use a CRL (certificate revocation list) instead (see the
<tt class="docutils literal"><span class="pre">--crl-verify</span></tt> option).</p>
<p class="last">This option must be associated with a specific client instance, which
means that it must be specified either in a client instance config file
using <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> or dynamically generated using a
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> script.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--connect-freq <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Allow a maximum of <tt class="docutils literal">n</tt> new connections per <tt class="docutils literal">sec</tt> seconds from
clients.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
connect-freq n sec
</pre>
<p>This is designed to contain DoS attacks which flood the server
with connection requests using certificates which will ultimately fail
to authenticate.</p>
<p>This limit applies after <tt class="docutils literal"><span class="pre">--connect-freq-initial</span></tt> and
only applies to client that have completed the three-way handshake
or client that use <tt class="docutils literal"><span class="pre">--tls-crypt-v2</span></tt> without cookie support
(<tt class="docutils literal"><span class="pre">allow-noncookie</span></tt> argument to <tt class="docutils literal"><span class="pre">--tls-crypt-v2</span></tt>).</p>
<p>This is an imperfect solution however, because in a real DoS scenario,
legitimate connections might also be refused.</p>
<p class="last">For the best protection against DoS attacks in server mode, use
<tt class="docutils literal"><span class="pre">--proto</span> udp</tt> and either <tt class="docutils literal"><span class="pre">--tls-auth</span></tt> or <tt class="docutils literal"><span class="pre">--tls-crypt</span></tt>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--connect-freq-initial <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">(UDP only) Allow a maximum of <tt class="docutils literal">n</tt> initial connection packet responses
per <tt class="docutils literal">sec</tt> seconds from the OpenVPN server to clients.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
connect-freq-initial n sec
</pre>
<p>OpenVPN starting at 2.6 is very efficient in responding to initial
connection packets. When not limiting the initial responses
an OpenVPN daemon can be abused in reflection attacks.
This option is designed to limit the rate OpenVPN will respond to initial
attacks.</p>
<p class="last">Connection attempts that complete the initial three-way handshake
will not be counted against the limit. The default is to allow
100 initial connection per 10s.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--duplicate-cn</span></kbd></td>
<td>Allow multiple clients with the same common name to concurrently
connect. In the absence of this option, OpenVPN will disconnect a client
instance upon connection of a new client having the same common name.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ifconfig-pool <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Set aside a pool of subnets to be dynamically allocated to connecting
clients, similar to a DHCP server.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
ifconfig-pool start-IP end-IP [netmask]
</pre>
<p class="last">For tun-style tunnels, each client
will be given a /30 subnet (for interoperability with Windows clients).
For tap-style tunnels, individual addresses will be allocated, and the
optional <tt class="docutils literal">netmask</tt> parameter will also be pushed to clients.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ifconfig-ipv6-pool <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Specify an IPv6 address pool for dynamic assignment to clients.</p>
<p>Valid args:</p>
<pre class="literal-block">
ifconfig-ipv6-pool ipv6addr/bits
</pre>
<p class="last">The pool starts at <tt class="docutils literal">ipv6addr</tt> and matches the offset determined from
the start of the IPv4 pool. If the host part of the given IPv6
address is <tt class="docutils literal">0</tt>, the pool starts at <tt class="docutils literal">ipv6addr</tt> +1.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ifconfig-pool-persist <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Persist/unpersist ifconfig-pool data to <tt class="docutils literal">file</tt>, at <tt class="docutils literal">seconds</tt>
intervals (default <code>600</code>), as well as on program startup and shutdown.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
ifconfig-pool-persist file [seconds]
</pre>
<p>The goal of this option is to provide a long-term association between
clients (denoted by their common name) and the virtual IP address
assigned to them from the ifconfig-pool. Maintaining a long-term
association is good for clients because it allows them to effectively
use the <tt class="docutils literal"><span class="pre">--persist-tun</span></tt> option.</p>
<p><tt class="docutils literal">file</tt> is a comma-delimited ASCII file, formatted as
<code><Common-Name>,<IP-address></code>.</p>
<p>If <tt class="docutils literal">seconds</tt> = <code>0</code>, <tt class="docutils literal">file</tt> will be treated as read-only. This
is useful if you would like to treat <tt class="docutils literal">file</tt> as a configuration file.</p>
<p class="last">Note that the entries in this file are treated by OpenVPN as
<em>suggestions</em> only, based on past associations between a common name and
IP address. They do not guarantee that the given common name will always
receive the given IP address. If you want guaranteed assignment, use
<tt class="docutils literal"><span class="pre">--ifconfig-push</span></tt></p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ifconfig-push <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Push virtual IP endpoints for client tunnel, overriding the
<tt class="docutils literal"><span class="pre">--ifconfig-pool</span></tt> dynamic allocation.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
ifconfig-push local remote-netmask [alias]
</pre>
<p>The parameters <tt class="docutils literal">local</tt> and <tt class="docutils literal"><span class="pre">remote-netmask</span></tt> are set according to the
<tt class="docutils literal"><span class="pre">--ifconfig</span></tt> directive which you want to execute on the client machine
to configure the remote end of the tunnel. Note that the parameters
<tt class="docutils literal">local</tt> and <tt class="docutils literal"><span class="pre">remote-netmask</span></tt> are from the perspective of the client,
not the server. They may be DNS names rather than IP addresses, in which
case they will be resolved on the server at the time of client
connection.</p>
<p>The optional <tt class="docutils literal">alias</tt> parameter may be used in cases where NAT causes
the client view of its local endpoint to differ from the server view. In
this case <tt class="docutils literal"><span class="pre">local/remote-netmask</span></tt> will refer to the server view while
<tt class="docutils literal"><span class="pre">alias/remote-netmask</span></tt> will refer to the client view.</p>
<p>This option must be associated with a specific client instance, which
means that it must be specified either in a client instance config file
using <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> or dynamically generated using a
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> script.</p>
<p>Remember also to include a <tt class="docutils literal"><span class="pre">--route</span></tt> directive in the main OpenVPN
config file which encloses <tt class="docutils literal">local</tt>, so that the kernel will know to
route it to the server's TUN/TAP interface.</p>
<p>OpenVPN's internal client IP address selection algorithm works as
follows:</p>
<ol class="last arabic simple">
<li>Use <tt class="docutils literal"><span class="pre">--client-connect</span> script</tt> generated file for static IP
(first choice).</li>
<li>Use <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> file for static IP (next choice).</li>
<li>Use <tt class="docutils literal"><span class="pre">--ifconfig-pool</span></tt> allocation for dynamic IP (last
choice).</li>
</ol>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ifconfig-ipv6-push <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">for <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> per-client static IPv6 interface
configuration, see <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> and <tt class="docutils literal"><span class="pre">--ifconfig-push</span></tt> for
more details.</p>
<p>Valid syntax:</p>
<pre class="last literal-block">
ifconfig-ipv6-push ipv6addr/bits ipv6remote
</pre>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--multihome</span></kbd></td>
<td><p class="first">Configure a multi-homed UDP server. This option needs to be used when a
server has more than one IP address (e.g. multiple interfaces, or
secondary IP addresses), and is not using <tt class="docutils literal"><span class="pre">--local</span></tt> to force binding
to one specific address only. This option will add some extra lookups to
the packet path to ensure that the UDP reply packets are always sent
from the address that the client is talking to. This is not supported on
all platforms, and it adds more processing, so it's not enabled by
default.</p>
<dl class="last docutils">
<dt><em>Notes:</em></dt>
<dd><ul class="first last simple">
<li>This option is only relevant for UDP servers.</li>
<li>If you do an IPv6+IPv4 dual-stack bind on a Linux machine with
multiple IPv4 address, connections to IPv4 addresses will not
work right on kernels before 3.15, due to missing kernel
support for the IPv4-mapped case (some distributions have
ported this to earlier kernel versions, though).</li>
</ul>
</dd>
</dl>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--iroute <var>args</var></span></kbd></td>
<td><p class="first">Generate an internal route to a specific client. The <tt class="docutils literal">netmask</tt>
parameter, if omitted, defaults to <code>255.255.255.255</code>.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
iroute network [netmask]
</pre>
<p>This directive can be used to route a fixed subnet from the server to a
particular client, regardless of where the client is connecting from.
Remember that you must also add the route to the system routing table as
well (such as by using the <tt class="docutils literal"><span class="pre">--route</span></tt> directive). The reason why two
routes are needed is that the <tt class="docutils literal"><span class="pre">--route</span></tt> directive routes the packet
from the kernel to OpenVPN. Once in OpenVPN, the <tt class="docutils literal"><span class="pre">--iroute</span></tt> directive
routes to the specific client.</p>
<p>However, when using DCO, the <tt class="docutils literal"><span class="pre">--iroute</span></tt> directive is usually enough
for DCO to fully configure the routing table. The extra <tt class="docutils literal"><span class="pre">--route</span></tt>
directive is required only if the expected behaviour is to route the
traffic for a specific network to the VPN interface also when the
responsible client is not connected (traffic will then be dropped).</p>
<p>This option must be specified either in a client instance config file
using <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> or dynamically generated using a
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> script.</p>
<p class="last">The <tt class="docutils literal"><span class="pre">--iroute</span></tt> directive also has an important interaction with
<tt class="docutils literal"><span class="pre">--push</span> "route <span class="pre">..."</span></tt>. <tt class="docutils literal"><span class="pre">--iroute</span></tt> essentially defines a subnet which
is owned by a particular client (we will call this client <em>A</em>). If you
would like other clients to be able to reach <em>A</em>'s subnet, you can use
<tt class="docutils literal"><span class="pre">--push</span> "route <span class="pre">..."</span></tt> together with <tt class="docutils literal"><span class="pre">--client-to-client</span></tt> to effect
this. In order for all clients to see <em>A</em>'s subnet, OpenVPN must push
this route to all clients EXCEPT for <em>A</em>, since the subnet is already
owned by <em>A</em>. OpenVPN accomplishes this by not not pushing a route to
a client if it matches one of the client's iroutes.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--iroute-ipv6 <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">for <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> per-client static IPv6 route configuration,
see <tt class="docutils literal"><span class="pre">--iroute</span></tt> for more details how to setup and use this, and how
<tt class="docutils literal"><span class="pre">--iroute</span></tt> and <tt class="docutils literal"><span class="pre">--route</span></tt> interact.</p>
<p>Valid syntax:</p>
<pre class="last literal-block">
iroute-ipv6 ipv6addr/bits
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--max-clients <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td>Limit server to a maximum of <tt class="docutils literal">n</tt> concurrent clients.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--max-routes-per-client <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Allow a maximum of <tt class="docutils literal">n</tt> internal routes per client (default
<code>256</code>). This is designed to help contain DoS attacks where an
authenticated client floods the server with packets appearing to come
from many unique MAC addresses, forcing the server to deplete virtual
memory as its internal routing table expands. This directive can be used
in a <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> file or auto-generated by a
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> script to override the global value for a particular
client.</p>
<p class="last">Note that this directive affects OpenVPN's internal routing table, not
the kernel routing table.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--opt-verify</span></kbd></td>
<td><p class="first"><strong>DEPRECATED</strong> Clients that connect with options that are incompatible with
those of the server will be disconnected.</p>
<p>Options that will be compared for compatibility include <tt class="docutils literal"><span class="pre">dev-type</span></tt>,
<tt class="docutils literal"><span class="pre">link-mtu</span></tt>, <tt class="docutils literal"><span class="pre">tun-mtu</span></tt>, <tt class="docutils literal">proto</tt>, <tt class="docutils literal">ifconfig</tt>,
<tt class="docutils literal"><span class="pre">comp-lzo</span></tt>, <tt class="docutils literal">fragment</tt>, <tt class="docutils literal">keydir</tt>, <tt class="docutils literal">cipher</tt>,
<tt class="docutils literal">auth</tt>, <tt class="docutils literal">keysize</tt>, <tt class="docutils literal">secret</tt>, <tt class="docutils literal"><span class="pre">no-replay</span></tt>,
<tt class="docutils literal"><span class="pre">tls-auth</span></tt>, <tt class="docutils literal"><span class="pre">key-method</span></tt>, <tt class="docutils literal"><span class="pre">tls-server</span></tt>
and <tt class="docutils literal"><span class="pre">tls-client</span></tt>.</p>
<p class="last">This option requires that <tt class="docutils literal"><span class="pre">--disable-occ</span></tt> NOT be used.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--port-share <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Share OpenVPN TCP with another service</p>
<p>Valid syntax:</p>
<pre class="literal-block">
port-share host port [dir]
</pre>
<p>When run in TCP server mode, share the OpenVPN port with another
application, such as an HTTPS server. If OpenVPN senses a connection to
its port which is using a non-OpenVPN protocol, it will proxy the
connection to the server at <tt class="docutils literal">host</tt>:<tt class="docutils literal">port</tt>. Currently only designed to
work with HTTP/HTTPS, though it would be theoretically possible to
extend to other protocols such as ssh.</p>
<p><tt class="docutils literal">dir</tt> specifies an optional directory where a temporary file with name
N containing content C will be dynamically generated for each proxy
connection, where N is the source IP:port of the client connection and C
is the source IP:port of the connection to the proxy receiver. This
directory can be used as a dictionary by the proxy receiver to determine
the origin of the connection. Each generated file will be automatically
deleted when the proxied connection is torn down.</p>
<p class="last">Not implemented on Windows.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--push <var>option</var></span></kbd></td>
<td><p class="first">Push a config file option back to the client for remote execution. Note
that <tt class="docutils literal">option</tt> must be enclosed in double quotes (<code>""</code>). The
client must specify <tt class="docutils literal"><span class="pre">--pull</span></tt> in its config file. The set of options
which can be pushed is limited by both feasibility and security. Some
options such as those which would execute scripts are banned, since they
would effectively allow a compromised server to execute arbitrary code
on the client. Other options such as TLS or MTU parameters cannot be
pushed because the client needs to know them before the connection to the
server can be initiated.</p>
<p class="last">This is a partial list of options which can currently be pushed:
<tt class="docutils literal"><span class="pre">--route</span></tt>, <tt class="docutils literal"><span class="pre">--route-gateway</span></tt>, <tt class="docutils literal"><span class="pre">--route-delay</span></tt>,
<tt class="docutils literal"><span class="pre">--redirect-gateway</span></tt>, <tt class="docutils literal"><span class="pre">--ip-win32</span></tt>, <tt class="docutils literal"><span class="pre">--dhcp-option</span></tt>, <tt class="docutils literal"><span class="pre">--dns</span></tt>,
<tt class="docutils literal"><span class="pre">--inactive</span></tt>, <tt class="docutils literal"><span class="pre">--ping</span></tt>, <tt class="docutils literal"><span class="pre">--ping-exit</span></tt>, <tt class="docutils literal"><span class="pre">--ping-restart</span></tt>,
<tt class="docutils literal"><span class="pre">--setenv</span></tt>, <tt class="docutils literal"><span class="pre">--auth-token</span></tt>, <tt class="docutils literal"><span class="pre">--persist-key</span></tt>, <tt class="docutils literal"><span class="pre">--persist-tun</span></tt>,
<tt class="docutils literal"><span class="pre">--echo</span></tt>, <tt class="docutils literal"><span class="pre">--comp-lzo</span></tt>, <tt class="docutils literal"><span class="pre">--socket-flags</span></tt>, <tt class="docutils literal"><span class="pre">--sndbuf</span></tt>,
<tt class="docutils literal"><span class="pre">--rcvbuf</span></tt>, <tt class="docutils literal"><span class="pre">--session-timeout</span></tt></p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--push-remove <var>opt</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Selectively remove all <tt class="docutils literal"><span class="pre">--push</span></tt> options matching "opt" from the option
list for a client. <tt class="docutils literal">opt</tt> is matched as a substring against the whole
option string to-be-pushed to the client, so <tt class="docutils literal"><span class="pre">--push-remove</span> route</tt>
would remove all <tt class="docutils literal"><span class="pre">--push</span> route ...</tt> and <tt class="docutils literal"><span class="pre">--push</span> <span class="pre">route-ipv6</span> ...</tt>
statements, while <tt class="docutils literal"><span class="pre">--push-remove</span> <span class="pre">"route-ipv6</span> 2001:"</tt> would only remove
IPv6 routes for <code>2001:...</code> networks.</p>
<p><tt class="docutils literal"><span class="pre">--push-remove</span></tt> can only be used in a client-specific context, like in
a <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> file, or <tt class="docutils literal"><span class="pre">--client-connect</span></tt> script or plugin
-- similar to <tt class="docutils literal"><span class="pre">--push-reset</span></tt>, just more selective.</p>
<p><em>NOTE</em>: to <em>change</em> an option, <tt class="docutils literal"><span class="pre">--push-remove</span></tt> can be used to first
remove the old value, and then add a new <tt class="docutils literal"><span class="pre">--push</span></tt> option with the new
value.</p>
<p class="last"><em>NOTE 2</em>: due to implementation details, 'ifconfig' and 'ifconfig-ipv6'
can only be removed with an exact match on the option (
<code>push-remove ifconfig</code>), no substring matching and no matching on
the IPv4/IPv6 address argument is possible.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--push-reset</span></kbd></td>
<td><p class="first">Don't inherit the global push list for a specific client instance.
Specify this option in a client-specific context such as with a
<tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> configuration file. This option will ignore
<tt class="docutils literal"><span class="pre">--push</span></tt> options at the global config file level.</p>
<p class="last"><em>NOTE</em>: <tt class="docutils literal"><span class="pre">--push-reset</span></tt> is very thorough: it will remove almost
all options from the list of to-be-pushed options. In many cases,
some of these options will need to be re-configured afterwards -
specifically, <tt class="docutils literal"><span class="pre">--topology</span> subnet</tt> and <tt class="docutils literal"><span class="pre">--route-gateway</span></tt> will get
lost and this will break client configs in many cases. Thus, for most
purposes, <tt class="docutils literal"><span class="pre">--push-remove</span></tt> is better suited to selectively remove
push options for individual clients.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--server <var>args</var></span></kbd></td>
<td><p class="first">A helper directive designed to simplify the configuration of OpenVPN's
server mode. This directive will set up an OpenVPN server which will
allocate addresses to clients out of the given network/netmask. The
server itself will take the <code>.1</code> address of the given network for
use as the server-side endpoint of the local TUN/TAP interface. If the
optional <code>nopool</code> flag is given, no dynamic IP address pool will
prepared for VPN clients.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
server network netmask [nopool]
</pre>
<p>For example, <tt class="docutils literal"><span class="pre">--server</span> 10.8.0.0 255.255.255.0</tt> expands as follows:</p>
<pre class="literal-block">
mode server
tls-server
push "topology [topology]"
if dev tun AND (topology == net30 OR topology == p2p):
ifconfig 10.8.0.1 10.8.0.2
if !nopool:
ifconfig-pool 10.8.0.4 10.8.0.251
route 10.8.0.0 255.255.255.0
if client-to-client:
push "route 10.8.0.0 255.255.255.0"
else if topology == net30:
push "route 10.8.0.1"
if dev tap OR (dev tun AND topology == subnet):
ifconfig 10.8.0.1 255.255.255.0
if !nopool:
ifconfig-pool 10.8.0.2 10.8.0.253 255.255.255.0
push "route-gateway 10.8.0.1"
if route-gateway unset:
route-gateway 10.8.0.2
</pre>
<p class="last">Don't use <tt class="docutils literal"><span class="pre">--server</span></tt> if you are ethernet bridging. Use
<tt class="docutils literal"><span class="pre">--server-bridge</span></tt> instead.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--server-bridge <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">A helper directive similar to <tt class="docutils literal"><span class="pre">--server</span></tt> which is designed to simplify
the configuration of OpenVPN's server mode in ethernet bridging
configurations.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
server-bridge gateway netmask pool-start-IP pool-end-IP
server-bridge [nogw]
</pre>
<p>If <tt class="docutils literal"><span class="pre">--server-bridge</span></tt> is used without any parameters, it will enable a
DHCP-proxy mode, where connecting OpenVPN clients will receive an IP
address for their TAP adapter from the DHCP server running on the
OpenVPN server-side LAN. Note that only clients that support the binding
of a DHCP client with the TAP adapter (such as Windows) can support this
mode. The optional <code>nogw</code> flag (advanced) indicates that gateway
information should not be pushed to the client.</p>
<p>To configure ethernet bridging, you must first use your OS's bridging
capability to bridge the TAP interface with the ethernet NIC interface.
For example, on Linux this is done with the <code>brctl</code> tool, and with
Windows XP it is done in the Network Connections Panel by selecting the
ethernet and TAP adapters and right-clicking on "Bridge Connections".</p>
<p>Next you you must manually set the IP/netmask on the bridge interface.
The <tt class="docutils literal">gateway</tt> and <tt class="docutils literal">netmask</tt> parameters to <tt class="docutils literal"><span class="pre">--server-bridge</span></tt> can be
set to either the IP/netmask of the bridge interface, or the IP/netmask
of the default gateway/router on the bridged subnet.</p>
<p>Finally, set aside a IP range in the bridged subnet, denoted by
<tt class="docutils literal"><span class="pre">pool-start-IP</span></tt> and <tt class="docutils literal"><span class="pre">pool-end-IP</span></tt>, for OpenVPN to allocate to
connecting clients.</p>
<p>For example, <tt class="docutils literal"><span class="pre">server-bridge</span> 10.8.0.4 255.255.255.0 10.8.0.128
10.8.0.254</tt> expands as follows:</p>
<pre class="literal-block">
mode server
tls-server
ifconfig-pool 10.8.0.128 10.8.0.254 255.255.255.0
push "route-gateway 10.8.0.4"
</pre>
<p>In another example, <tt class="docutils literal"><span class="pre">--server-bridge</span></tt> (without parameters) expands as
follows:</p>
<pre class="literal-block">
mode server
tls-server
push "route-gateway dhcp"
</pre>
<p>Or <tt class="docutils literal"><span class="pre">--server-bridge</span> nogw</tt> expands as follows:</p>
<pre class="last literal-block">
mode server
tls-server
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--server-ipv6 <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Convenience-function to enable a number of IPv6 related options at once,
namely <tt class="docutils literal"><span class="pre">--ifconfig-ipv6</span></tt>, <tt class="docutils literal"><span class="pre">--ifconfig-ipv6-pool</span></tt> and
<tt class="docutils literal"><span class="pre">--push</span> <span class="pre">tun-ipv6</span></tt>.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
server-ipv6 ipv6addr/bits
</pre>
<p class="last">Pushing of the <tt class="docutils literal"><span class="pre">--tun-ipv6</span></tt> directive is done for older clients which
require an explicit <tt class="docutils literal"><span class="pre">--tun-ipv6</span></tt> in their configuration.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--stale-routes-check <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Remove routes which haven't had activity for <tt class="docutils literal">n</tt> seconds (i.e. the ageing
time). This check is run every <tt class="docutils literal">t</tt> seconds (i.e. check interval).</p>
<p>Valid syntax:</p>
<pre class="literal-block">
stale-routes-check n [t]
</pre>
<p>If <tt class="docutils literal">t</tt> is not present it defaults to <tt class="docutils literal">n</tt>.</p>
<p class="last">This option helps to keep the dynamic routing table small. See also
<tt class="docutils literal"><span class="pre">--max-routes-per-client</span></tt></p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--username-as-common-name</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Use the authenticated username as the common-name, rather than the
common-name from the client certificate. Requires that some form of
<tt class="docutils literal"><span class="pre">--auth-user-pass</span></tt> verification is in effect. As the replacement happens
after <tt class="docutils literal"><span class="pre">--auth-user-pass</span></tt> verification, the verification script or
plugin will still receive the common-name from the certificate.</p>
<p class="last">The common_name environment variable passed to scripts and plugins invoked
after authentication (e.g, client-connect script) and file names parsed in
client-config directory will match the username.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--verify-client-cert <var>mode</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Specify whether the client is required to supply a valid certificate.</p>
<p>Possible <tt class="docutils literal">mode</tt> options are:</p>
<dl class="docutils">
<dt><code>none</code></dt>
<dd><p class="first">A client certificate is not required. the client needs to
authenticate using username/password only. Be aware that using this
directive is less secure than requiring certificates from all
clients.</p>
<p>If you use this directive, the entire responsibility of authentication
will rest on your <tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> script, so keep in mind
that bugs in your script could potentially compromise the security of
your VPN.</p>
<p class="last"><tt class="docutils literal"><span class="pre">--verify-client-cert</span> none</tt> is functionally equivalent to
<tt class="docutils literal"><span class="pre">--client-cert-not-required</span></tt>.</p>
</dd>
<dt><code>optional</code></dt>
<dd><p class="first">A client may present a certificate but it is not required to do so.
When using this directive, you should also use a
<tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> script to ensure that clients are
authenticated using a certificate, a username and password, or
possibly even both.</p>
<p class="last">Again, the entire responsibility of authentication will rest on your
<tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> script, so keep in mind that bugs in your
script could potentially compromise the security of your VPN.</p>
</dd>
<dt><code>require</code></dt>
<dd>This is the default option. A client is required to present a
certificate, otherwise VPN access is refused.</dd>
</dl>
<p class="last">If you don't use this directive (or use <tt class="docutils literal"><span class="pre">--verify-client-cert</span> require</tt>)
but you also specify an <tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> script, then OpenVPN
will perform double authentication. The client certificate verification
AND the <tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> script will need to succeed in order
for a client to be authenticated and accepted onto the VPN.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--vlan-tagging</span></kbd></td>
<td><p class="first">Server-only option. Turns the OpenVPN server instance into a switch that
understands VLAN-tagging, based on IEEE 802.1Q.</p>
<p>The server TAP device and each of the connecting clients is seen as a
port of the switch. All client ports are in untagged mode and the server
TAP device is VLAN-tagged, untagged or accepts both, depending on the
<tt class="docutils literal"><span class="pre">--vlan-accept</span></tt> setting.</p>
<p>Ethernet frames with a prepended 802.1Q tag are called "tagged". If the
VLAN Identifier (VID) field in such a tag is non-zero, the frame is
called "VLAN-tagged". If the VID is zero, but the Priority Control Point
(PCP) field is non-zero, the frame is called "prio-tagged". If there is
no 802.1Q tag, the frame is "untagged".</p>
<p>Using the <tt class="docutils literal"><span class="pre">--vlan-pvid</span> v</tt> option once per client (see
--client-config-dir), each port can be associated with a certain VID.
Packets can only be forwarded between ports having the same VID.
Therefore, clients with differing VIDs are completely separated from
one-another, even if <tt class="docutils literal"><span class="pre">--client-to-client</span></tt> is activated.</p>
<p>The packet filtering takes place in the OpenVPN server. Clients should
not have any VLAN tagging configuration applied.</p>
<p>The <tt class="docutils literal"><span class="pre">--vlan-tagging</span></tt> option is off by default. While turned off,
OpenVPN accepts any Ethernet frame and does not perform any special
processing for VLAN-tagged packets.</p>
<p class="last">This option can only be activated in <tt class="docutils literal"><span class="pre">--dev</span> tap mode</tt>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--vlan-accept <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Configure the VLAN tagging policy for the server TAP device.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
vlan-accept all|tagged|untagged
</pre>
<p>The following modes are available:</p>
<dl class="docutils">
<dt><code>tagged</code></dt>
<dd>Admit only VLAN-tagged frames. Only VLAN-tagged packets are accepted,
while untagged or priority-tagged packets are dropped when entering
the server TAP device.</dd>
<dt><code>untagged</code></dt>
<dd>Admit only untagged and prio-tagged frames. VLAN-tagged packets are
not accepted, while untagged or priority-tagged packets entering the
server TAP device are tagged with the value configured for the global
<tt class="docutils literal"><span class="pre">--vlan-pvid</span></tt> setting.</dd>
<dt><code>all</code> (default)</dt>
<dd>Admit all frames. All packets are admitted and then treated like
untagged or tagged mode respectively.</dd>
<dt><em>Note</em>:</dt>
<dd>Some vendors refer to switch ports running in <code>tagged</code> mode
as "trunk ports" and switch ports running in <code>untagged</code> mode
as "access ports".</dd>
</dl>
<p>Packets forwarded from clients to the server are VLAN-tagged with the
originating client's PVID, unless the VID matches the global
<tt class="docutils literal"><span class="pre">--vlan-pvid</span></tt>, in which case the tag is removed.</p>
<p class="last">If no <em>PVID</em> is configured for a given client (see --vlan-pvid) packets
are tagged with 1 by default.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--vlan-pvid <var>v</var></span></kbd></td>
<td><p class="first">Specifies which VLAN identifier a "port" is associated with. Only valid
when <tt class="docutils literal"><span class="pre">--vlan-tagging</span></tt> is specified.</p>
<p>In the client context, the setting specifies which VLAN ID a client is
associated with. In the global context, the VLAN ID of the server TAP
device is set. The latter only makes sense for <tt class="docutils literal"><span class="pre">--vlan-accept</span>
untagged</tt> and <tt class="docutils literal"><span class="pre">--vlan-accept</span> all</tt> modes.</p>
<p>Valid values for <tt class="docutils literal">v</tt> go from <code>1</code> through to <code>4094</code>. The
global value defaults to <code>1</code>. If no <tt class="docutils literal"><span class="pre">--vlan-pvid</span></tt> is specified in
the client context, the global value is inherited.</p>
<p class="last">In some switch implementations, the <em>PVID</em> is also referred to as "Native
VLAN".</p>
</td></tr>
</tbody>
</table>
</div>
</div>
<div class="section" id="encryption-options">
<h1>Encryption Options</h1>
<div class="section" id="ssl-library-information">
<h2>SSL Library information</h2>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group">
<kbd><span class="option">--show-ciphers</span></kbd></td>
<td>(Standalone) Show all cipher algorithms to use with the <tt class="docutils literal"><span class="pre">--cipher</span></tt>
option.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--show-digests</span></kbd></td>
<td>(Standalone) Show all message digest algorithms to use with the
<tt class="docutils literal"><span class="pre">--auth</span></tt> option.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--show-tls</span></kbd></td>
<td><p class="first">(Standalone) Show all TLS ciphers supported by the crypto library.
OpenVPN uses TLS to secure the control channel, over which the keys that
are used to protect the actual VPN traffic are exchanged. The TLS
ciphers will be sorted from highest preference (most secure) to lowest.</p>
<p class="last">Be aware that whether a cipher suite in this list can actually work
depends on the specific setup of both peers (e.g. both peers must
support the cipher, and an ECDSA cipher suite will not work if you are
using an RSA certificate, etc.).</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--show-engines</span></kbd></td>
<td>(Standalone) Show currently available hardware-based crypto acceleration
engines supported by the OpenSSL library.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--show-groups</span></kbd></td>
<td>(Standalone) Show all available elliptic curves/groups to use with the
<tt class="docutils literal"><span class="pre">--ecdh-curve</span></tt> and <tt class="docutils literal"><span class="pre">tls-groups</span></tt> options.</td></tr>
</tbody>
</table>
</div>
<div class="section" id="generating-key-material">
<h2>Generating key material</h2>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group">
<kbd><span class="option">--genkey <var>args</var></span></kbd></td>
<td><p class="first">(Standalone) Generate a key to be used of the type keytype. if keyfile
is left out or empty the key will be output on stdout. See the following
sections for the different keytypes.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
--genkey keytype keyfile
</pre>
<p>Valid keytype arguments are:</p>
<p><code>secret</code> Standard OpenVPN shared secret keys</p>
<p><code>tls-crypt</code> Alias for <code>secret</code></p>
<p><code>tls-auth</code> Alias for <code>secret</code></p>
<p><code>auth-token</code> Key used for <tt class="docutils literal"><span class="pre">--auth-gen-token-key</span></tt></p>
<p><code>tls-crypt-v2-server</code> TLS Crypt v2 server key</p>
<p><code>tls-crypt-v2-client</code> TLS Crypt v2 client key</p>
<p>Examples:</p>
<pre class="literal-block">
$ openvpn --genkey secret shared.key
$ openvpn --genkey tls-crypt shared.key
$ openvpn --genkey tls-auth shared.key
$ openvpn --genkey tls-crypt-v2-server v2crypt-server.key
$ openvpn --tls-crypt-v2 v2crypt-server.key --genkey tls-crypt-v2-client v2crypt-client-1.key
</pre>
<ul class="last">
<li><p class="first">Generating <em>Shared Secret Keys</em>
Generate a shared secret, for use with the <tt class="docutils literal"><span class="pre">--secret</span></tt>, <tt class="docutils literal"><span class="pre">--tls-auth</span></tt>
or <tt class="docutils literal"><span class="pre">--tls-crypt</span></tt> options.</p>
<p>Syntax:</p>
<pre class="literal-block">
$ openvpn --genkey secret|tls-crypt|tls-auth keyfile
</pre>
<p>The key is saved in <tt class="docutils literal">keyfile</tt>. All three variants (<tt class="docutils literal"><span class="pre">--secret</span></tt>,
<tt class="docutils literal"><span class="pre">tls-crypt</span></tt> and <tt class="docutils literal"><span class="pre">tls-auth</span></tt>) generate the same type of key. The
aliases are added for convenience.</p>
<p>If using this for <tt class="docutils literal"><span class="pre">--secret</span></tt>, this file must be shared with the peer
over a pre-existing secure channel such as <tt class="docutils literal">scp</tt>(1).</p>
</li>
<li><p class="first">Generating <em>TLS Crypt v2 Server key</em>
Generate a <tt class="docutils literal"><span class="pre">--tls-crypt-v2</span></tt> key to be used by an OpenVPN server.
The key is stored in <tt class="docutils literal">keyfile</tt>.</p>
<p>Syntax:</p>
<pre class="literal-block">
--genkey tls-crypt-v2-server keyfile
</pre>
</li>
<li><p class="first">Generating <em>TLS Crypt v2 Client key</em>
Generate a --tls-crypt-v2 key to be used by OpenVPN clients. The
key is stored in <tt class="docutils literal">keyfile</tt>.</p>
<p>Syntax</p>
<pre class="literal-block">
--genkey tls-crypt-v2-client keyfile [metadata]
</pre>
<p>If supplied, include the supplied <tt class="docutils literal">metadata</tt> in the wrapped client
key. This metadata must be supplied in base64-encoded form. The
metadata must be at most 733 bytes long (980 characters in base64, though
note that 980 base64 characters can encode more than 733 bytes).</p>
<p>If no metadata is supplied, OpenVPN will use a 64-bit unix timestamp
representing the current time in UTC, encoded in network order, as
metadata for the generated key.</p>
<p>A tls-crypt-v2 client key is wrapped using a server key. To generate a
client key, the user must therefore supply the server key using the
<tt class="docutils literal"><span class="pre">--tls-crypt-v2</span></tt> option.</p>
<p>Servers can use <tt class="docutils literal"><span class="pre">--tls-crypt-v2-verify</span></tt> to specify a metadata
verification command.</p>
</li>
<li><p class="first">Generate <em>Authentication Token key</em>
Generate a new secret that can be used with <strong>--auth-gen-token-secret</strong></p>
<p>Syntax:</p>
<pre class="literal-block">
--genkey auth-token [keyfile]
</pre>
<dl class="docutils">
<dt><em>Note:</em></dt>
<dd><p class="first last">This file should be kept secret to the server as anyone that has
access to this file will be able to generate auth tokens that the
OpenVPN server will accept as valid.</p>
</dd>
</dl>
</li>
</ul>
</td></tr>
</tbody>
</table>
</div>
<div class="section" id="data-channel-renegotiation">
<h2>Data Channel Renegotiation</h2>
<p>When running OpenVPN in client/server mode, the data channel will use a
separate ephemeral encryption key which is rotated at regular intervals.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--reneg-bytes <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Renegotiate data channel key after <tt class="docutils literal">n</tt> bytes sent or received
(disabled by default with an exception, see below). OpenVPN allows the
lifetime of a key to be expressed as a number of bytes
encrypted/decrypted, a number of packets, or a number of seconds. A key
renegotiation will be forced if any of these three criteria are met by
either peer.</p>
<p class="last">If using ciphers with cipher block sizes less than 128-bits,
<tt class="docutils literal"><span class="pre">--reneg-bytes</span></tt> is set to 64MB by default, unless it is explicitly
disabled by setting the value to <code>0</code>, but this is
<strong>HIGHLY DISCOURAGED</strong> as this is designed to add some protection against
the SWEET32 attack vector. For more information see the <tt class="docutils literal"><span class="pre">--cipher</span></tt>
option.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--reneg-pkts <var>n</var></span></kbd></td>
<td>Renegotiate data channel key after <strong>n</strong> packets sent and received
(disabled by default).</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--reneg-sec <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Renegotiate data channel key after at most <tt class="docutils literal">max</tt> seconds
(default <code>3600</code>) and at least <tt class="docutils literal">min</tt> seconds (default is 90% of
<tt class="docutils literal">max</tt> for servers, and equal to <tt class="docutils literal">max</tt> for clients).</p>
<pre class="literal-block">
reneg-sec max [min]
</pre>
<p>The effective <tt class="docutils literal"><span class="pre">--reneg-sec</span></tt> value used is per session
pseudo-uniform-randomized between <tt class="docutils literal">min</tt> and <tt class="docutils literal">max</tt>.</p>
<p>With the default value of <code>3600</code> this results in an effective per
session value in the range of <code>3240</code> .. <code>3600</code> seconds for
servers, or just 3600 for clients.</p>
<p>When using dual-factor authentication, note that this default value may
cause the end user to be challenged to reauthorize once per hour.</p>
<p class="last">Also, keep in mind that this option can be used on both the client and
server, and whichever uses the lower value will be the one to trigger
the renegotiation. A common mistake is to set <tt class="docutils literal"><span class="pre">--reneg-sec</span></tt> to a
higher value on either the client or server, while the other side of the
connection is still using the default value of <code>3600</code> seconds,
meaning that the renegotiation will still occur once per <code>3600</code>
seconds. The solution is to increase --reneg-sec on both the client and
server, or set it to <code>0</code> on one side of the connection (to
disable), and to your chosen value on the other side.</p>
</td></tr>
</tbody>
</table>
</div>
<div class="section" id="tls-mode-options">
<h2>TLS Mode Options</h2>
<p>TLS mode is the most powerful crypto mode of OpenVPN in both security
and flexibility. TLS mode works by establishing control and data
channels which are multiplexed over a single TCP/UDP port. OpenVPN
initiates a TLS session over the control channel and uses it to exchange
cipher and HMAC keys to protect the data channel. TLS mode uses a robust
reliability layer over the UDP connection for all control channel
communication, while the data channel, over which encrypted tunnel data
passes, is forwarded without any mediation. The result is the best of
both worlds: a fast data channel that forwards over UDP with only the
overhead of encrypt, decrypt, and HMAC functions, and a control channel
that provides all of the security features of TLS, including
certificate-based authentication and Diffie Hellman forward secrecy.</p>
<p>To use TLS mode, each peer that runs OpenVPN should have its own local
certificate/key pair (<tt class="docutils literal"><span class="pre">--cert</span></tt> and <tt class="docutils literal"><span class="pre">--key</span></tt>), signed by the root
certificate which is specified in <tt class="docutils literal"><span class="pre">--ca</span></tt>.</p>
<p>When two OpenVPN peers connect, each presents its local certificate to
the other. Each peer will then check that its partner peer presented a
certificate which was signed by the master root certificate as specified
in <tt class="docutils literal"><span class="pre">--ca</span></tt>.</p>
<p>If that check on both peers succeeds, then the TLS negotiation will
succeed, both OpenVPN peers will exchange temporary session keys, and
the tunnel will begin passing data.</p>
<p>The OpenVPN project provides a set of scripts for managing RSA
certificates and keys: <a class="reference external" href="https://github.com/OpenVPN/easy-rsa">https://github.com/OpenVPN/easy-rsa</a></p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group">
<kbd><span class="option">--askpass <var>file</var></span></kbd></td>
<td><p class="first">Get certificate password from console or <tt class="docutils literal">file</tt> before we daemonize.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
askpass
askpass file
</pre>
<p>For the extremely security conscious, it is possible to protect your
private key with a password. Of course this means that every time the
OpenVPN daemon is started you must be there to type the password. The
<tt class="docutils literal"><span class="pre">--askpass</span></tt> option allows you to start OpenVPN from the command line.
It will query you for a password before it daemonizes. To protect a
private key with a password you should omit the <tt class="docutils literal"><span class="pre">-nodes</span></tt> option when
you use the <tt class="docutils literal">openssl</tt> command line tool to manage certificates and
private keys.</p>
<p class="last">If <tt class="docutils literal">file</tt> is specified, read the password from the first line of
<tt class="docutils literal">file</tt>. Keep in mind that storing your password in a file to a certain
extent invalidates the extra security provided by using an encrypted
key.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--ca <var>file</var></span></kbd></td>
<td><p class="first">Certificate authority (CA) file in .pem format, also referred to as the
<em>root</em> certificate. This file can have multiple certificates in .pem
format, concatenated together. You can construct your own certificate
authority certificate and private key by using a command such as:</p>
<pre class="literal-block">
openssl req -nodes -new -x509 -keyout ca.key -out ca.crt
</pre>
<p>Then edit your openssl.cnf file and edit the <tt class="docutils literal">certificate</tt> variable to
point to your new root certificate <tt class="docutils literal">ca.crt</tt>.</p>
<p class="last">For testing purposes only, the OpenVPN distribution includes a sample CA
certificate (ca.crt). Of course you should never use the test
certificates and test keys distributed with OpenVPN in a production
environment, since by virtue of the fact that they are distributed with
OpenVPN, they are totally insecure.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--capath <var>dir</var></span></kbd></td>
<td><p class="first">Directory containing trusted certificates (CAs and CRLs). Not available
with mbed TLS.</p>
<p>CAs in the capath directory are expected to be named <hash>.<n>. CRLs
are expected to be named <hash>.r<n>. See the <tt class="docutils literal"><span class="pre">-CApath</span></tt> option of
<tt class="docutils literal">openssl verify</tt>, and the <tt class="docutils literal"><span class="pre">-hash</span></tt> option of <tt class="docutils literal">openssl x509</tt>,
<tt class="docutils literal">openssl crl</tt> and <tt class="docutils literal">X509_LOOKUP_hash_dir()</tt>(3)
for more information.</p>
<p class="last">Similar to the <tt class="docutils literal"><span class="pre">--crl-verify</span></tt> option, CRLs are not mandatory -
OpenVPN will log the usual warning in the logs if the relevant CRL is
missing, but the connection will be allowed.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--cert <var>file</var></span></kbd></td>
<td><p class="first">Local peer's signed certificate in .pem format -- must be signed by a
certificate authority whose certificate is in <tt class="docutils literal"><span class="pre">--ca</span> file</tt>. Each peer
in an OpenVPN link running in TLS mode should have its own certificate
and private key file. In addition, each certificate should have been
signed by the key of a certificate authority whose public key resides in
the <tt class="docutils literal"><span class="pre">--ca</span></tt> certificate authority file. You can easily make your own
certificate authority (see above) or pay money to use a commercial
service such as thawte.com (in which case you will be helping to finance
the world's second space tourist :). To generate a certificate, you can
use a command such as:</p>
<pre class="literal-block">
openssl req -nodes -new -keyout mycert.key -out mycert.csr
</pre>
<p>If your certificate authority private key lives on another machine, copy
the certificate signing request (mycert.csr) to this other machine (this
can be done over an insecure channel such as email). Now sign the
certificate with a command such as:</p>
<pre class="literal-block">
openssl ca -out mycert.crt -in mycert.csr
</pre>
<p class="last">Now copy the certificate (mycert.crt) back to the peer which initially
generated the .csr file (this can be over a public medium). Note that
the <tt class="docutils literal">openssl ca</tt> command reads the location of the certificate
authority key from its configuration file such as
<code>/usr/share/ssl/openssl.cnf</code> -- note also that for certificate
authority functions, you must set up the files <code>index.txt</code> (may be
empty) and <code>serial</code> (initialize to <code>01</code>).</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--crl-verify <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Check peer certificate against a Certificate Revocation List.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
crl-verify file/directory flag
</pre>
<p>Examples:</p>
<pre class="literal-block">
crl-verify crl-file.pem
crl-verify /etc/openvpn/crls dir
</pre>
<p>A CRL (certificate revocation list) is used when a particular key is
compromised but when the overall PKI is still intact.</p>
<p>Suppose you had a PKI consisting of a CA, root certificate, and a number
of client certificates. Suppose a laptop computer containing a client
key and certificate was stolen. By adding the stolen certificate to the
CRL file, you could reject any connection which attempts to use it,
while preserving the overall integrity of the PKI.</p>
<p>The only time when it would be necessary to rebuild the entire PKI from
scratch would be if the root certificate key itself was compromised.</p>
<p>The option is not mandatory - if the relevant CRL is missing, OpenVPN
will log a warning in the logs - e.g.</p>
<pre class="literal-block">
VERIFY WARNING: depth=0, unable to get certificate CRL
</pre>
<p>but the connection will be allowed. If the optional <code>dir</code> flag
is specified, enable a different mode where the <tt class="docutils literal"><span class="pre">crl-verify</span></tt> is
pointed at a directory containing files named as revoked serial numbers
(the files may be empty, the contents are never read). If a client
requests a connection, where the client certificate serial number
(decimal string) is the name of a file present in the directory, it will
be rejected.</p>
<dl class="last docutils">
<dt><em>Note:</em></dt>
<dd>As the crl file (or directory) is read every time a peer
connects, if you are dropping root privileges with
<tt class="docutils literal"><span class="pre">--user</span></tt>, make sure that this user has sufficient
privileges to read the file.</dd>
</dl>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--dh <var>file</var></span></kbd></td>
<td><p class="first">File containing Diffie Hellman parameters in .pem format (required for
<tt class="docutils literal"><span class="pre">--tls-server</span></tt> only).</p>
<p>Set <tt class="docutils literal">file</tt> to <code>none</code> to disable Diffie Hellman key exchange (and
use ECDH only). Note that this requires peers to be using an SSL library
that supports ECDH TLS cipher suites (e.g. OpenSSL 1.0.1+, or
mbed TLS 2.0+).</p>
<p class="last">Use <tt class="docutils literal">openssl dhparam <span class="pre">-out</span> dh2048.pem 2048</tt> to generate 2048-bit DH
parameters. Diffie Hellman parameters may be considered public.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ecdh-curve <var>name</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Specify the curve to use for elliptic curve Diffie Hellman. Available
curves can be listed with <tt class="docutils literal"><span class="pre">--show-curves</span></tt>. The specified curve will
only be used for ECDH TLS-ciphers.</p>
<p class="last">This option is not supported in mbed TLS builds of OpenVPN.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--extra-certs <var>file</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Specify a <tt class="docutils literal">file</tt> containing one or more PEM certs (concatenated
together) that complete the local certificate chain.</p>
<p class="last">This option is useful for "split" CAs, where the CA for server certs is
different than the CA for client certs. Putting certs in this file
allows them to be used to complete the local certificate chain without
trusting them to verify the peer-submitted certificate, as would be the
case if the certs were placed in the <tt class="docutils literal">ca</tt> file.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--hand-window <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Handshake Window -- the TLS-based key exchange must finalize within
<tt class="docutils literal">n</tt> seconds of handshake initiation by any peer (default <code>60</code>
seconds). If the handshake fails we will attempt to reset our connection
with our peer and try again. Even in the event of handshake failure we
will still use our expiring key for up to <tt class="docutils literal"><span class="pre">--tran-window</span></tt> seconds to
maintain continuity of transmission of tunnel data.</p>
<p class="last">The <tt class="docutils literal"><span class="pre">--hand-window</span></tt> parameter also controls the amount of time that
the OpenVPN client repeats the pull request until it times out.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--key <var>file</var></span></kbd></td>
<td>Local peer's private key in .pem format. Use the private key which was
generated when you built your peer's certificate (see <tt class="docutils literal"><span class="pre">--cert</span> file</tt>
above).</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--pkcs12 <var>file</var></span></kbd></td>
<td>Specify a PKCS #12 file containing local private key, local certificate,
and root CA certificate. This option can be used instead of <tt class="docutils literal"><span class="pre">--ca</span></tt>,
<tt class="docutils literal"><span class="pre">--cert</span></tt>, and <tt class="docutils literal"><span class="pre">--key</span></tt>. Not available with mbed TLS.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--remote-cert-eku <var>oid</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Require that peer certificate was signed with an explicit <em>extended key
usage</em>.</p>
<p>This is a useful security option for clients, to ensure that the host
they connect to is a designated server.</p>
<p class="last">The extended key usage should be encoded in <em>oid notation</em>, or <em>OpenSSL
symbolic representation</em>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--remote-cert-ku <var>key-usage</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Require that peer certificate was signed with an explicit
<tt class="docutils literal"><span class="pre">key-usage</span></tt>.</p>
<p>If present in the certificate, the <code>keyUsage</code> value is validated by
the TLS library during the TLS handshake. Specifying this option without
arguments requires this extension to be present (so the TLS library will
verify it).</p>
<p>If <tt class="docutils literal"><span class="pre">key-usage</span></tt> is a list of usage bits, the <code>keyUsage</code> field
must have <em>at least</em> the same bits set as the bits in <em>one of</em> the values
supplied in the <tt class="docutils literal"><span class="pre">key-usage</span></tt> list.</p>
<p>The <tt class="docutils literal"><span class="pre">key-usage</span></tt> values in the list must be encoded in hex, e.g.</p>
<pre class="last literal-block">
remote-cert-ku a0
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--remote-cert-tls <var>type</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Require that peer certificate was signed with an explicit <em>key usage</em>
and <em>extended key usage</em> based on RFC3280 TLS rules.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
remote-cert-tls server
remote-cert-tls client
</pre>
<p>This is a useful security option for clients, to ensure that the host
they connect to is a designated server. Or the other way around; for a
server to verify that only hosts with a client certificate can connect.</p>
<p>The <tt class="docutils literal"><span class="pre">--remote-cert-tls</span> client</tt> option is equivalent to</p>
<pre class="literal-block">
remote-cert-ku
remote-cert-eku "TLS Web Client Authentication"
</pre>
<p>The <tt class="docutils literal"><span class="pre">--remote-cert-tls</span> server</tt> option is equivalent to</p>
<pre class="literal-block">
remote-cert-ku
remote-cert-eku "TLS Web Server Authentication"
</pre>
<p class="last">This is an important security precaution to protect against a
man-in-the-middle attack where an authorized client attempts to connect
to another client by impersonating the server. The attack is easily
prevented by having clients verify the server certificate using any one
of <tt class="docutils literal"><span class="pre">--remote-cert-tls</span></tt>, <tt class="docutils literal"><span class="pre">--verify-x509-name</span></tt>, <tt class="docutils literal"><span class="pre">--peer-fingerprint</span></tt>
or <tt class="docutils literal"><span class="pre">--tls-verify</span></tt>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-auth <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Add an additional layer of HMAC authentication on top of the TLS control
channel to mitigate DoS attacks and attacks on the TLS stack.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
tls-auth file
tls-auth file 0
tls-auth file 1
</pre>
<p>In a nutshell, <tt class="docutils literal"><span class="pre">--tls-auth</span></tt> enables a kind of "HMAC firewall" on
OpenVPN's TCP/UDP port, where TLS control channel packets bearing an
incorrect HMAC signature can be dropped immediately without response.</p>
<p><tt class="docutils literal">file</tt> (required) is a file in OpenVPN static key format which can be
generated by <tt class="docutils literal"><span class="pre">--genkey</span></tt>.</p>
<p>Older versions (up to OpenVPN 2.3) supported a freeform passphrase file.
This is no longer supported in newer versions (v2.4+).</p>
<p>See the <tt class="docutils literal"><span class="pre">--secret</span></tt> option for more information on the optional
<tt class="docutils literal">direction</tt> parameter.</p>
<p><tt class="docutils literal"><span class="pre">--tls-auth</span></tt> is recommended when you are running OpenVPN in a mode
where it is listening for packets from any IP address, such as when
<tt class="docutils literal"><span class="pre">--remote</span></tt> is not specified, or <tt class="docutils literal"><span class="pre">--remote</span></tt> is specified with
<tt class="docutils literal"><span class="pre">--float</span></tt>.</p>
<p>The rationale for this feature is as follows. TLS requires a
multi-packet exchange before it is able to authenticate a peer. During
this time before authentication, OpenVPN is allocating resources (memory
and CPU) to this potential peer. The potential peer is also exposing
many parts of OpenVPN and the OpenSSL library to the packets it is
sending. Most successful network attacks today seek to either exploit
bugs in programs (such as buffer overflow attacks) or force a program to
consume so many resources that it becomes unusable. Of course the first
line of defense is always to produce clean, well-audited code. OpenVPN
has been written with buffer overflow attack prevention as a top
priority. But as history has shown, many of the most widely used network
applications have, from time to time, fallen to buffer overflow attacks.</p>
<p>So as a second line of defense, OpenVPN offers this special layer of
authentication on top of the TLS control channel so that every packet on
the control channel is authenticated by an HMAC signature and a unique
ID for replay protection. This signature will also help protect against
DoS (Denial of Service) attacks. An important rule of thumb in reducing
vulnerability to DoS attacks is to minimize the amount of resources a
potential, but as yet unauthenticated, client is able to consume.</p>
<p><tt class="docutils literal"><span class="pre">--tls-auth</span></tt> does this by signing every TLS control channel packet
with an HMAC signature, including packets which are sent before the TLS
level has had a chance to authenticate the peer. The result is that
packets without the correct signature can be dropped immediately upon
reception, before they have a chance to consume additional system
resources such as by initiating a TLS handshake. <tt class="docutils literal"><span class="pre">--tls-auth</span></tt> can be
strengthened by adding the <tt class="docutils literal"><span class="pre">--replay-persist</span></tt> option which will keep
OpenVPN's replay protection state in a file so that it is not lost
across restarts.</p>
<p>It should be emphasized that this feature is optional and that the key
file used with <tt class="docutils literal"><span class="pre">--tls-auth</span></tt> gives a peer nothing more than the power
to initiate a TLS handshake. It is not used to encrypt or authenticate
any tunnel data.</p>
<p class="last">Use <tt class="docutils literal"><span class="pre">--tls-crypt</span></tt> instead if you want to use the key file to not only
authenticate, but also encrypt the TLS control channel.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-groups <var>list</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">A list of allowable groups/curves in order of preference.</p>
<p>Set the allowed elliptic curves/groups for the TLS session.
These groups are allowed to be used in signatures and key exchange.</p>
<p>mbedTLS currently allows all known curves per default.</p>
<p>OpenSSL 1.1+ restricts the list per default to</p>
<pre class="literal-block">
"X25519:secp256r1:X448:secp521r1:secp384r1".
</pre>
<p>If you use certificates that use non-standard curves, you
might need to add them here. If you do not force the ecdh curve
by using <tt class="docutils literal"><span class="pre">--ecdh-curve</span></tt>, the groups for ecdh will also be picked
from this list.</p>
<p>OpenVPN maps the curve name <cite>secp256r1</cite> to <cite>prime256v1</cite> to allow
specifying the same tls-groups option for mbedTLS and OpenSSL.</p>
<p class="last">Warning: this option not only affects elliptic curve certificates
but also the key exchange in TLS 1.3 and using this option improperly
will disable TLS 1.3.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-cert-profile <var>profile</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Set the allowed cryptographic algorithms for certificates according to
<tt class="docutils literal">profile</tt>.</p>
<p>The following profiles are supported:</p>
<dl class="docutils">
<dt><code>insecure</code></dt>
<dd>Identical for mbed TLS to <cite>legacy</cite></dd>
<dt><code>legacy</code> (default)</dt>
<dd>SHA1 and newer, RSA 2048-bit+, any elliptic curve.</dd>
<dt><code>preferred</code></dt>
<dd>SHA2 and newer, RSA 2048-bit+, any elliptic curve.</dd>
<dt><code>suiteb</code></dt>
<dd>SHA256/SHA384, ECDSA with P-256 or P-384.</dd>
</dl>
<p>This option is only fully supported for mbed TLS builds. OpenSSL builds
use the following approximation:</p>
<dl class="docutils">
<dt><code>insecure</code></dt>
<dd>sets "security level 0"</dd>
<dt><code>legacy</code> (default)</dt>
<dd>sets "security level 1"</dd>
<dt><code>preferred</code></dt>
<dd>sets "security level 2"</dd>
<dt><code>suiteb</code></dt>
<dd>sets "security level 3" and <tt class="docutils literal"><span class="pre">--tls-cipher</span> "SUITEB128"</tt>.</dd>
</dl>
<p class="last">OpenVPN will migrate to 'preferred' as default in the future. Please
ensure that your keys already comply.</p>
</td></tr>
</tbody>
</table>
<dl class="docutils">
<dt><em>WARNING:</em> <tt class="docutils literal"><span class="pre">--tls-ciphers</span></tt>, <tt class="docutils literal"><span class="pre">--tls-ciphersuites</span></tt> and <tt class="docutils literal"><span class="pre">tls-groups</span></tt></dt>
<dd>These options are expert features, which - if used correctly - can
improve the security of your VPN connection. But it is also easy to
unwittingly use them to carefully align a gun with your foot, or just
break your connection. Use with care!</dd>
</dl>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group">
<kbd><span class="option">--tls-cipher <var>l</var></span></kbd></td>
<td><p class="first">A list <tt class="docutils literal">l</tt> of allowable TLS ciphers delimited by a colon ("<code>:</code>").</p>
<p>These setting can be used to ensure that certain cipher suites are used
(or not used) for the TLS connection. OpenVPN uses TLS to secure the
control channel, over which the keys that are used to protect the actual
VPN traffic are exchanged.</p>
<p>The supplied list of ciphers is (after potential OpenSSL/IANA name
translation) simply supplied to the crypto library. Please see the
OpenSSL and/or mbed TLS documentation for details on the cipher list
interpretation.</p>
<p>For OpenSSL, the <tt class="docutils literal"><span class="pre">--tls-cipher</span></tt> is used for TLS 1.2 and below.</p>
<p>Use <tt class="docutils literal"><span class="pre">--show-tls</span></tt> to see a list of TLS ciphers supported by your crypto
library.</p>
<p class="last">The default for <tt class="docutils literal"><span class="pre">--tls-cipher</span></tt> is to use mbed TLS's default cipher list
when using mbed TLS or
<code>DEFAULT:!EXP:!LOW:!MEDIUM:!kDH:!kECDH:!DSS:!PSK:!SRP:!kRSA</code> when
using OpenSSL.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-ciphersuites <var>l</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Same as <tt class="docutils literal"><span class="pre">--tls-cipher</span></tt> but for TLS 1.3 and up. mbed TLS has no
TLS 1.3 support yet and only the <tt class="docutils literal"><span class="pre">--tls-cipher</span></tt> setting is used.</p>
<p class="last">The default for <cite>--tls-ciphersuites</cite> is to use the crypto library's
default.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--tls-client</span></kbd></td>
<td>Enable TLS and assume client role during TLS handshake.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-crypt <var>keyfile</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Encrypt and authenticate all control channel packets with the key from
<tt class="docutils literal">keyfile</tt>. (See <tt class="docutils literal"><span class="pre">--tls-auth</span></tt> for more background.)</p>
<p>Encrypting (and authenticating) control channel packets:</p>
<ul class="simple">
<li>provides more privacy by hiding the certificate used for the TLS
connection,</li>
<li>makes it harder to identify OpenVPN traffic as such,</li>
<li>provides "poor-man's" post-quantum security, against attackers who will
never know the pre-shared key (i.e. no forward secrecy).</li>
</ul>
<p>In contrast to <tt class="docutils literal"><span class="pre">--tls-auth</span></tt>, <tt class="docutils literal"><span class="pre">--tls-crypt</span></tt> does <em>not</em> require the
user to set <tt class="docutils literal"><span class="pre">--key-direction</span></tt>.</p>
<p><strong>Security Considerations</strong></p>
<p>All peers use the same <tt class="docutils literal"><span class="pre">--tls-crypt</span></tt> pre-shared group key to
authenticate and encrypt control channel messages. To ensure that IV
collisions remain unlikely, this key should not be used to encrypt more
than 2^48 client-to-server or 2^48 server-to-client control channel
messages. A typical initial negotiation is about 10 packets in each
direction. Assuming both initial negotiation and renegotiations are at
most 2^16 (65536) packets (to be conservative), and (re)negotiations
happen each minute for each user (24/7), this limits the tls-crypt key
lifetime to 8171 years divided by the number of users. So a setup with
1000 users should rotate the key at least once each eight years. (And a
setup with 8000 users each year.)</p>
<p>If IV collisions were to occur, this could result in the security of
<tt class="docutils literal"><span class="pre">--tls-crypt</span></tt> degrading to the same security as using <tt class="docutils literal"><span class="pre">--tls-auth</span></tt>.
That is, the control channel still benefits from the extra protection
against active man-in-the-middle-attacks and DoS attacks, but may no
longer offer extra privacy and post-quantum security on top of what TLS
itself offers.</p>
<p class="last">For large setups or setups where clients are not trusted, consider using
<tt class="docutils literal"><span class="pre">--tls-crypt-v2</span></tt> instead. That uses per-client unique keys, and
thereby improves the bounds to 'rotate a client key at least once per
8000 years'.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-crypt-v2 <var>keyfile</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Valid syntax:</p>
<pre class="literal-block">
tls-crypt-v2 keyfile
tls-crypt-v2 keyfile force-cookie
tls-crypt-v2 keyfile allow-noncookie
</pre>
<p>Use client-specific tls-crypt keys.</p>
<p>For clients, <tt class="docutils literal">keyfile</tt> is a client-specific tls-crypt key. Such a key
can be generated using the <code>--genkey tls-crypt-v2-client</code> option.</p>
<p>For servers, <tt class="docutils literal">keyfile</tt> is used to unwrap client-specific keys supplied
by the client during connection setup. This key must be the same as the
key used to generate the client-specific key (see <code>--genkey
tls-crypt-v2-client</code>).</p>
<p>On servers, this option can be used together with the <tt class="docutils literal"><span class="pre">--tls-auth</span></tt> or
<tt class="docutils literal"><span class="pre">--tls-crypt</span></tt> option. In that case, the server will detect whether the
client is using client-specific keys, and automatically select the right
mode.</p>
<p class="last">The optional parameters <code>force-cookie</code> allows only tls-crypt-v2
clients that support a cookie based stateless three way handshake that
avoids replay attacks and state exhaustion on the server side (OpenVPN
2.6 and later). The option <code>allow-noncookie</code> explicitly allows
older tls-crypt-v2 clients. The default is (currently)
<code>allow-noncookie</code>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-crypt-v2-verify <var>cmd</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Run command <tt class="docutils literal">cmd</tt> to verify the metadata of the client-specific
tls-crypt-v2 key of a connecting client. This allows server
administrators to reject client connections, before exposing the TLS
stack (including the notoriously dangerous X.509 and ASN.1 stacks) to
the connecting client.</p>
<p>OpenVPN supplies the following environment variables to the command (and
only these variables. The normal environment variables available for
other scripts are NOT present):</p>
<ul class="simple">
<li><code>script_type</code> is set to <code>tls-crypt-v2-verify</code></li>
<li><code>metadata_type</code> is set to <code>0</code> if the metadata was user
supplied, or <code>1</code> if it's a 64-bit unix timestamp representing
the key creation time.</li>
<li><code>metadata_file</code> contains the filename of a temporary file that
contains the client metadata.</li>
</ul>
<p class="last">The command can reject the connection by exiting with a non-zero exit
code.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--tls-exit</span></kbd></td>
<td>Exit on TLS negotiation failure. This option can be useful when you only
want to make one attempt at connecting, e.g. in a test or monitoring script.
(OpenVPN's own test suite uses it this way.)</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--tls-server</span></kbd></td>
<td>Enable TLS and assume server role during TLS handshake. Note that
OpenVPN is designed as a peer-to-peer application. The designation of
client or server is only for the purpose of negotiating the TLS control
channel.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-timeout <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td>Packet retransmit timeout on TLS control channel if no acknowledgment
from remote within <tt class="docutils literal">n</tt> seconds (default <code>2</code>). When OpenVPN sends
a control packet to its peer, it will expect to receive an
acknowledgement within <tt class="docutils literal">n</tt> seconds or it will retransmit the packet,
subject to a TCP-like exponential backoff algorithm. This parameter only
applies to control channel packets. Data channel packets (which carry
encrypted tunnel data) are never acknowledged, sequenced, or
retransmitted by OpenVPN because the higher level network protocols
running on top of the tunnel such as TCP expect this role to be left to
them.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-version-min <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Sets the minimum TLS version we will accept from the peer (default in
2.6.0 and later is "1.2").</p>
<p>Valid syntax:</p>
<pre class="literal-block">
tls-version-min version ['or-highest']
</pre>
<p class="last">Examples for version include <code>1.0</code>, <code>1.1</code>, or <code>1.2</code>. If
<code>or-highest</code> is specified and version is not recognized, we will
only accept the highest TLS version supported by the local SSL
implementation.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-version-max <var>version</var></span></kbd></td>
</tr>
<tr><td> </td><td>Set the maximum TLS version we will use (default is the highest version
supported). Examples for version include <code>1.0</code>, <code>1.1</code>, or
<code>1.2</code>.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--verify-hash <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first"><strong>DEPRECATED</strong> Specify SHA1 or SHA256 fingerprint for level-1 cert.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
verify-hash hash [algo]
</pre>
<p>The level-1 cert is the CA (or intermediate cert) that signs the leaf
certificate, and is one removed from the leaf certificate in the
direction of the root. When accepting a connection from a peer, the
level-1 cert fingerprint must match <tt class="docutils literal">hash</tt> or certificate verification
will fail. Hash is specified as XX:XX:... For example:</p>
<pre class="literal-block">
AD:B0:95:D8:09:C8:36:45:12:A9:89:C8:90:09:CB:13:72:A6:AD:16
</pre>
<p>The <tt class="docutils literal">algo</tt> flag can be either <code>SHA1</code> or <code>SHA256</code>. If not
provided, it defaults to <code>SHA1</code>.</p>
<p>This option can also be inlined</p>
<pre class="last literal-block">
<verify-hash>
00:11:22:33:44:55:66:77:88:99:aa:bb:cc:dd:ee:ff:00:11:22:33:44:55:66:77:88:99:aa:bb:cc:dd:ee:ff
11:22:33:44:55:66:77:88:99:aa:bb:cc:dd:ee:ff:00:11:22:33:44:55:66:77:88:99:aa:bb:cc:dd:ee:ff:00
</verify-hash>
</pre>
</td></tr>
</tbody>
</table>
<p>If the option is inlined, <tt class="docutils literal">algo</tt> is always <code>SHA256</code>.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--peer-fingerprint <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><blockquote class="first">
Specify a SHA256 fingerprint or list of SHA256 fingerprints to verify
the peer certificate against. The peer certificate must match one of the
fingerprint or certificate verification will fail. The option can also
be inlined</blockquote>
<p>Valid syntax:</p>
<pre class="literal-block">
peer-fingerprint AD:B0:95:D8:09:...
</pre>
<p>or inline:</p>
<pre class="literal-block">
<peer-fingerprint>
00:11:22:33:44:55:66:77:88:99:aa:bb:cc:dd:ee:ff:00:11:22:33:44:55:66:77:88:99:aa:bb:cc:dd:ee:ff
11:22:33:44:55:66:77:88:99:aa:bb:cc:dd:ee:ff:00:11:22:33:44:55:66:77:88:99:aa:bb:cc:dd:ee:ff:00
</peer-fingerprint>
</pre>
<p class="last">When the <tt class="docutils literal"><span class="pre">--peer-fingerprint</span></tt> option is used, specifying a CA with <tt class="docutils literal"><span class="pre">--ca</span></tt> or <tt class="docutils literal"><span class="pre">--capath</span></tt> is
optional. This allows the he <tt class="docutils literal"><span class="pre">--peer-fingerprint</span></tt> to be used as alternative to a PKI with
self-signed certificates for small setups. See the examples section for such a setup.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--verify-x509-name <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Accept connections only if a host's X.509 name is equal to <strong>name.</strong> The
remote host must also pass all other tests of verification.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
verify-x509 name type
</pre>
<p>Which X.509 name is compared to <tt class="docutils literal">name</tt> depends on the setting of type.
<tt class="docutils literal">type</tt> can be <code>subject</code> to match the complete subject DN
(default), <code>name</code> to match a subject RDN or <code>name-prefix</code> to
match a subject RDN prefix. Which RDN is verified as name depends on the
<tt class="docutils literal"><span class="pre">--x509-username-field</span></tt> option. But it defaults to the common name
(CN), e.g. a certificate with a subject DN</p>
<pre class="literal-block">
C=KG, ST=NA, L=Bishkek, CN=Server-1
</pre>
<p>would be matched by:</p>
<pre class="literal-block">
verify-x509-name 'C=KG, ST=NA, L=Bishkek, CN=Server-1'
verify-x509-name Server-1 name
verify-x509-name Server- name-prefix
</pre>
<p>The last example is useful if you want a client to only accept
connections to <code>Server-1</code>, <code>Server-2</code>, etc.</p>
<p><tt class="docutils literal"><span class="pre">--verify-x509-name</span></tt> is a useful replacement for the <tt class="docutils literal"><span class="pre">--tls-verify</span></tt>
option to verify the remote host, because <tt class="docutils literal"><span class="pre">--verify-x509-name</span></tt> works
in a <tt class="docutils literal"><span class="pre">--chroot</span></tt> environment without any dependencies.</p>
<p>Using a name prefix is a useful alternative to managing a CRL
(Certificate Revocation List) on the client, since it allows the client
to refuse all certificates except for those associated with designated
servers.</p>
<dl class="last docutils">
<dt><em>NOTE:</em></dt>
<dd>Test against a name prefix only when you are using OpenVPN
with a custom CA certificate that is under your control. Never use
this option with type <code>name-prefix</code> when your client
certificates are signed by a third party, such as a commercial
web CA.</dd>
</dl>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--x509-track <var>attribute</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Save peer X509 <strong>attribute</strong> value in environment for use by plugins and
management interface. Prepend a <code>+</code> to <tt class="docutils literal">attribute</tt> to save values
from full cert chain. Otherwise the attribute will only be exported for
the leaf cert (i.e. depth <code>0</code> of the cert chain). Values will be
encoded as <code>X509_<depth>_<attribute>=<value></code>. Multiple <tt class="docutils literal"><span class="pre">--x509-track</span></tt>
options can be defined to track multiple attributes.</p>
<p><tt class="docutils literal">attribute</tt> can be any part of the X509 Subject field or any X509v3
extension (RFC 3280). X509v3 extensions might not be supported when
not using the default TLS backend library (OpenSSL). You can also
request the <tt class="docutils literal">SHA1</tt> and <tt class="docutils literal">SHA256</tt> fingerprints of the cert,
but that is always exported as <code>tls_digest_{n}</code> and
<code>tls_digest_sha256_{n}</code> anyway.</p>
<p>Note that by default <strong>all</strong> parts of the X509 Subject field are exported in
the environment for the whole cert chain. If you use <tt class="docutils literal"><span class="pre">--x509-track</span></tt> at least
once <strong>only</strong> the attributes specified by these options are exported.</p>
<p>Examples:</p>
<pre class="last literal-block">
x509-track CN # exports only X509_0_CN
x509-track +CN # exports X509_{n}_CN for chain
x509-track basicConstraints # exports value of "X509v3 Basic Constraints"
x509-track SHA256 # exports SHA256 fingerprint
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--x509-username-field <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Fields in the X.509 certificate subject to be used as the username
(default <code>CN</code>). If multiple fields are specified their values
will be concatenated into the one username using <code>_</code> symbol
as a separator.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
x509-username-field [ext:]fieldname [[ext:]fieldname...]
</pre>
<p>Typically, this option is specified with <strong>fieldname</strong> arguments as
either of the following:</p>
<pre class="literal-block">
x509-username-field emailAddress
x509-username-field ext:subjectAltName
x509-username-field CN serialNumber
</pre>
<p>The first example uses the value of the <code>emailAddress</code> attribute
in the certificate's Subject field as the username. The second example
uses the <code>ext:</code> prefix to signify that the X.509 extension
<tt class="docutils literal">fieldname</tt> <code>subjectAltName</code> be searched for an rfc822Name
(email) field to be used as the username. In cases where there are
multiple email addresses in <code>ext:fieldname</code>, the last occurrence
is chosen. The last example uses the value of the <code>CN</code> attribute
in the Subject field, combined with the <code>_</code> separator and the
hexadecimal representation of the certificate's <code>serialNumber</code>.</p>
<p>When this option is used, the <tt class="docutils literal"><span class="pre">--verify-x509-name</span></tt> option will match
against the chosen <tt class="docutils literal">fieldname</tt> instead of the Common Name.</p>
<p>Only the <code>subjectAltName</code> and <code>issuerAltName</code> X.509
extensions and <code>serialNumber</code> X.509 attribute are supported.</p>
<p><strong>Please note:</strong> This option has a feature which will convert an
all-lowercase <tt class="docutils literal">fieldname</tt> to uppercase characters, e.g.,
<code>ou</code> -> <code>OU</code>. A mixed-case <tt class="docutils literal">fieldname</tt> or one having the
<code>ext:</code> prefix will be left as-is. This automatic upcasing feature is
deprecated and will be removed in a future release.</p>
<p class="last">Non-compliant symbols are being replaced with the <code>_</code> symbol, same as
the field separator, so concatenating multiple fields with such or <code>_</code>
symbols can potentially lead to username collisions.</p>
</td></tr>
</tbody>
</table>
</div>
<div class="section" id="pkcs-11-smartcard-options">
<h2>PKCS#11 / SmartCard options</h2>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--pkcs11-cert-private <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Set if access to certificate object should be performed after login.
Every provider has its own setting.</p>
<p>Valid syntaxes:</p>
<pre class="last literal-block">
pkcs11-cert-private 0
pkcs11-cert-private 1
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--pkcs11-id <var>name</var></span></kbd></td>
</tr>
<tr><td> </td><td>Specify the serialized certificate id to be used. The id can be gotten
by the standalone <tt class="docutils literal"><span class="pre">--show-pkcs11-ids</span></tt> option. See also the description
of <tt class="docutils literal"><span class="pre">--pkcs11-providers</span></tt> option.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--pkcs11-id-management</span></kbd></td>
</tr>
<tr><td> </td><td>Acquire PKCS#11 id from management interface. In this case a
<code>NEED-STR 'pkcs11-id-request'</code> real-time message will be triggered,
application may use pkcs11-id-count command to retrieve available number of
certificates, and pkcs11-id-get command to retrieve certificate id and
certificate body.
See also the description of <tt class="docutils literal"><span class="pre">--pkcs11-providers</span></tt> option.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--pkcs11-pin-cache <var>seconds</var></span></kbd></td>
</tr>
<tr><td> </td><td>Specify how many seconds the PIN can be cached, the default is until the
token is removed.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--pkcs11-private-mode <var>mode</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Specify which method to use in order to perform private key operations.
A different mode can be specified for each provider. Mode is encoded as
hex number, and can be a mask one of the following:</p>
<p><code>0</code> (default) Try to determine automatically.</p>
<p><code>1</code> Use sign.</p>
<p><code>2</code> Use sign recover.</p>
<p><code>4</code> Use decrypt.</p>
<p class="last"><code>8</code> Use unwrap.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--pkcs11-protected-authentication <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Use PKCS#11 protected authentication path, useful for biometric and
external keypad devices. Every provider has its own setting.</p>
<p>Valid syntaxes:</p>
<pre class="last literal-block">
pkcs11-protected-authentication 0
pkcs11-protected-authentication 1
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--pkcs11-providers <var>providers</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Specify an RSA Security Inc. PKCS #11 Cryptographic Token Interface
(Cryptoki) providers to load. A space-separated list of one or more
provider library names may be specified. This option along with <tt class="docutils literal"><span class="pre">--pkcs11-id</span></tt>
or <tt class="docutils literal"><span class="pre">pkcs11-id-management</span></tt> can be used instead of
<tt class="docutils literal"><span class="pre">--cert</span></tt> and <tt class="docutils literal"><span class="pre">--key</span></tt> or <tt class="docutils literal"><span class="pre">--pkcs12</span></tt>.</p>
<p class="last">If p11-kit is present on the system and was enabled during build, its
<code>p11-kit-proxy.so</code> module will be loaded by default if either
the <tt class="docutils literal"><span class="pre">--pkcs11-id</span></tt> or <tt class="docutils literal"><span class="pre">--pkcs11-id-management</span></tt> options is present without
<tt class="docutils literal"><span class="pre">--pkcs11-providers</span></tt>. If default loading is not enabled in the build and
no providers are specified, the former options will be ignored.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--show-pkcs11-ids <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">(Standalone) Show PKCS#11 token object list.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
show-pkcs11 [provider] [cert_private]
</pre>
<p>Specify <tt class="docutils literal">cert_private</tt> as <code>1</code> if certificates are stored as
private objects.</p>
<p>If <em>p11-kit</em> is present on the system, the <tt class="docutils literal">provider</tt> argument is
optional; if omitted the default <code>p11-kit-proxy.so</code> module will be
queried.</p>
<p class="last"><tt class="docutils literal"><span class="pre">--verb</span></tt> option can be used BEFORE this option to produce debugging
information.</p>
</td></tr>
</tbody>
</table>
</div>
</div>
<div class="section" id="data-channel-cipher-negotiation">
<h1>Data channel cipher negotiation</h1>
<p>OpenVPN 2.4 and higher have the capability to negotiate the data cipher that
is used to encrypt data packets. This section describes the mechanism in more detail and the
different backwards compatibility mechanism with older server and clients.</p>
<div class="section" id="openvpn-2-5-and-later-behaviour">
<h2>OpenVPN 2.5 and later behaviour</h2>
<p>When both client and server are at least running OpenVPN 2.5, that the order of
the ciphers of the server's <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> is used to pick the the data cipher.
That means that the first cipher in that list that is also in the client's
<tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> list is chosen. If no common cipher is found the client is rejected
with a AUTH_FAILED message (as seen in client log):</p>
<blockquote>
AUTH: Received control message: AUTH_FAILED,Data channel cipher negotiation failed (no shared cipher)</blockquote>
<p>OpenVPN 2.5 and later will only allow the ciphers specified in <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt>.
If <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> is not set the default is <code>AES-256-GCM:AES-128-GCM</code>.
In 2.6 and later the default is changed to
<code>AES-256-GCM:AES-128-GCM:CHACHA20-POLY1305</code> when Chacha20-Poly1305 is available.</p>
<p>For backwards compatibility OpenVPN 2.6 and later with <tt class="docutils literal"><span class="pre">--compat-mode</span> 2.4.x</tt>
(or lower) and OpenVPN 2.5 will automatically add a cipher specified using the
<tt class="docutils literal"><span class="pre">--cipher</span></tt> option to this list.</p>
</div>
<div class="section" id="openvpn-2-4-clients">
<h2>OpenVPN 2.4 clients</h2>
<p>The negotiation support in OpenVPN 2.4 was the first iteration of the implementation
and still had some quirks. Its main goal was "upgrade to AES-256-GCM when possible".
An OpenVPN 2.4 client that is built against a crypto library that supports AES in GCM
mode and does not have <tt class="docutils literal"><span class="pre">--ncp-disable</span></tt> will always announce support for
<cite>AES-256-GCM</cite> and <cite>AES-128-GCM</cite> to a server by sending <code>IV_NCP=2</code>.</p>
<p>This only causes a problem if <tt class="docutils literal"><span class="pre">--ncp-ciphers</span></tt> option has been changed from the
default of <code>AES-256-GCM:AES-128-GCM</code> to a value that does not include
these two ciphers. When an OpenVPN server tries to use <cite>AES-256-GCM</cite> or
<cite>AES-128-GCM</cite> the connection will then fail. It is therefore recommended to
always have the <cite>AES-256-GCM</cite> and <cite>AES-128-GCM</cite> ciphers to the <tt class="docutils literal"><span class="pre">--ncp-ciphers</span></tt>
options to avoid this behaviour.</p>
</div>
<div class="section" id="openvpn-3-clients">
<h2>OpenVPN 3 clients</h2>
<p>Clients based on the OpenVPN 3.x library (<a class="reference external" href="https://github.com/openvpn/openvpn3/">https://github.com/openvpn/openvpn3/</a>)
do not have a configurable <tt class="docutils literal"><span class="pre">--ncp-ciphers</span></tt> or <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> option. Newer
versions by default disable legacy AES-CBC, BF-CBC, and DES-CBC ciphers.
These clients will always announce support for all their supported AEAD ciphers
(<cite>AES-256-GCM</cite>, <cite>AES-128-GCM</cite> and in newer versions also <cite>Chacha20-Poly1305</cite>).</p>
<p>To support OpenVPN 3.x based clients at least one of these ciphers needs to be
included in the server's <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> option.</p>
</div>
<div class="section" id="openvpn-2-3-and-older-clients-and-clients-with-ncp-disable">
<h2>OpenVPN 2.3 and older clients (and clients with <tt class="docutils literal"><span class="pre">--ncp-disable</span></tt>)</h2>
<p>When a client without cipher negotiation support connects to a server the
cipher specified with the <tt class="docutils literal"><span class="pre">--cipher</span></tt> option in the client configuration
must be included in the <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> option of the server to allow
the client to connect. Otherwise the client will be sent the <tt class="docutils literal">AUTH_FAILED</tt>
message that indicates no shared cipher.</p>
<p>If the client is 2.3 or older and has been configured with the
<tt class="docutils literal"><span class="pre">--enable-small</span></tt> <code>./configure</code> argument, using
<tt class="docutils literal"><span class="pre">data-ciphers-fallback</span> cipher</tt> in the server config file with the explicit
cipher used by the client is necessary.</p>
</div>
<div class="section" id="openvpn-2-4-server">
<h2>OpenVPN 2.4 server</h2>
<p>When a client indicates support for <cite>AES-128-GCM</cite> and <cite>AES-256-GCM</cite>
(with <tt class="docutils literal">IV_NCP=2</tt>) an OpenVPN 2.4 server will send the first
cipher of the <tt class="docutils literal"><span class="pre">--ncp-ciphers</span></tt> to the OpenVPN client regardless of what
the cipher is. To emulate the behaviour of an OpenVPN 2.4 client as close
as possible and have compatibility to a setup that depends on this quirk,
adding <cite>AES-128-GCM</cite> and <cite>AES-256-GCM</cite> to the client's <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt>
option is required. OpenVPN 2.5+ will only announce the <tt class="docutils literal">IV_NCP=2</tt> flag if
those ciphers are present.</p>
</div>
<div class="section" id="openvpn-2-3-and-older-servers-and-servers-with-ncp-disable">
<h2>OpenVPN 2.3 and older servers (and servers with <tt class="docutils literal"><span class="pre">--ncp-disable</span></tt>)</h2>
<p>The cipher used by the server must be included in <tt class="docutils literal"><span class="pre">--data-ciphers</span></tt> to
allow the client connecting to a server without cipher negotiation
support.
(For compatibility OpenVPN 2.5 will also accept the cipher set with
<tt class="docutils literal"><span class="pre">--cipher</span></tt>)</p>
<p>If the server is 2.3 or older and has been configured with the
<tt class="docutils literal"><span class="pre">--enable-small</span></tt> <code>./configure</code> argument, adding
<tt class="docutils literal"><span class="pre">--data-ciphers-fallback</span> cipher</tt> to the client config with the explicit
cipher used by the server is necessary.</p>
</div>
<div class="section" id="blowfish-in-cbc-mode-bf-cbc-deprecation">
<h2>Blowfish in CBC mode (BF-CBC) deprecation</h2>
<p>The <tt class="docutils literal"><span class="pre">--cipher</span></tt> option defaulted to <cite>BF-CBC</cite> in OpenVPN 2.4 and older
version. The default was never changed to ensure backwards compatibility.
In OpenVPN 2.5 this behaviour has now been changed so that if the <tt class="docutils literal"><span class="pre">--cipher</span></tt>
is not explicitly set it does not allow the weak <cite>BF-CBC</cite> cipher any more
and needs to explicitly added as <tt class="docutils literal"><span class="pre">--cipher</span> <span class="pre">BFC-CBC</span></tt> or added to
<tt class="docutils literal"><span class="pre">--data-ciphers</span></tt>.</p>
<p>We strongly recommend to switching away from BF-CBC to a
more secure cipher as soon as possible instead.</p>
</div>
</div>
<div class="section" id="network-configuration">
<h1>NETWORK CONFIGURATION</h1>
<p>OpenVPN consists of two sides of network configuration. One side is the
<em>link</em> between the local and remote side, the other side is the <em>virtual
network adapter</em> (tun/tap device).</p>
<div class="section" id="link-options">
<h2>Link Options</h2>
<p>This link options section covers options related to the connection between
the local and the remote host.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--bind <var>keywords</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Bind to local address and port. This is the default unless any of
<tt class="docutils literal"><span class="pre">--proto</span> <span class="pre">tcp-client</span></tt> , <tt class="docutils literal"><span class="pre">--http-proxy</span></tt> or <tt class="docutils literal"><span class="pre">--socks-proxy</span></tt> are used.</p>
<p class="last">If the optional <code>ipv6only</code> keyword is present OpenVPN will bind only
to IPv6 (as opposed to IPv6 and IPv4) when a IPv6 socket is opened.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--float</span></kbd></td>
<td><p class="first">Allow remote peer to change its IP address and/or port number, such as
due to DHCP (this is the default if <tt class="docutils literal"><span class="pre">--remote</span></tt> is not used).
<tt class="docutils literal"><span class="pre">--float</span></tt> when specified with <tt class="docutils literal"><span class="pre">--remote</span></tt> allows an OpenVPN session
to initially connect to a peer at a known address, however if packets
arrive from a new address and pass all authentication tests, the new
address will take control of the session. This is useful when you are
connecting to a peer which holds a dynamic address such as a dial-in
user or DHCP client.</p>
<p class="last">Essentially, <tt class="docutils literal"><span class="pre">--float</span></tt> tells OpenVPN to accept authenticated packets
from any address, not only the address which was specified in the
<tt class="docutils literal"><span class="pre">--remote</span></tt> option.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--fragment <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Valid syntax:</p>
<pre class="literal-block">
fragment max
fragment max mtu
</pre>
<p>Enable internal datagram fragmentation so that no UDP datagrams are sent
which are larger than <tt class="docutils literal">max</tt> bytes.</p>
<p>If the <code>mtu</code> parameter is present the <tt class="docutils literal">max</tt> parameter is
interpreted to include IP and UDP encapsulation overhead. The
<code>mtu</code> parameter is introduced in OpenVPN version 2.6.0.</p>
<p>If the <code>mtu</code> parameter is absent, the <tt class="docutils literal">max</tt> parameter is
interpreted in the same way as the <tt class="docutils literal"><span class="pre">--link-mtu</span></tt> parameter, i.e.
the UDP packet size after encapsulation overhead has been added in,
but not including the UDP header itself.</p>
<p>The <tt class="docutils literal"><span class="pre">--fragment</span></tt> option only makes sense when you are using the UDP
protocol (<tt class="docutils literal"><span class="pre">--proto</span> udp</tt>).</p>
<p><tt class="docutils literal"><span class="pre">--fragment</span></tt> adds 4 bytes of overhead per datagram.</p>
<p>See the <tt class="docutils literal"><span class="pre">--mssfix</span></tt> option below for an important related option to
<tt class="docutils literal"><span class="pre">--fragment</span></tt>.</p>
<p>It should also be noted that this option is not meant to replace UDP
fragmentation at the IP stack level. It is only meant as a last resort
when path MTU discovery is broken. Using this option is less efficient
than fixing path MTU discovery for your IP link and using native IP
fragmentation instead.</p>
<p class="last">Having said that, there are circumstances where using OpenVPN's internal
fragmentation capability may be your only option, such as tunneling a
UDP multicast stream which requires fragmentation.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--keepalive <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">A helper directive designed to simplify the expression of <tt class="docutils literal"><span class="pre">--ping</span></tt> and
<tt class="docutils literal"><span class="pre">--ping-restart</span></tt>.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
keepalive interval timeout
</pre>
<p>Send ping once every <tt class="docutils literal">interval</tt> seconds, restart if ping is not received
for <tt class="docutils literal">timeout</tt> seconds.</p>
<p>This option can be used on both client and server side, but it is enough
to add this on the server side as it will push appropriate <tt class="docutils literal"><span class="pre">--ping</span></tt>
and <tt class="docutils literal"><span class="pre">--ping-restart</span></tt> options to the client. If used on both server and
client, the values pushed from server will override the client local
values.</p>
<p>The <tt class="docutils literal">timeout</tt> argument will be twice as long on the server side. This
ensures that a timeout is detected on client side before the server side
drops the connection.</p>
<p>For example, <tt class="docutils literal"><span class="pre">--keepalive</span> 10 60</tt> expands as follows:</p>
<pre class="last literal-block">
if mode server:
ping 10 # Argument: interval
ping-restart 120 # Argument: timeout*2
push "ping 10" # Argument: interval
push "ping-restart 60" # Argument: timeout
else
ping 10 # Argument: interval
ping-restart 60 # Argument: timeout
</pre>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--link-mtu <var>n</var></span></kbd></td>
<td><p class="first"><strong>DEPRECATED</strong> Sets an upper bound on the size of UDP packets which are sent between
OpenVPN peers. <em>It's best not to set this parameter unless you know what
you're doing.</em></p>
<p class="last">Due to variable header size of IP header (20 bytes for IPv4 and 40 bytes
for IPv6) and dynamically negotiated data channel cipher, this option
is not reliable. It is recommended to set tun-mtu with enough headroom
instead.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--local <var>host</var></span></kbd></td>
<td>Local host name or IP address for bind. If specified, OpenVPN will bind
to this address only. If unspecified, OpenVPN will bind to all
interfaces.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--lport <var>port</var></span></kbd></td>
<td>Set local TCP/UDP port number or name. Cannot be used together with
<tt class="docutils literal"><span class="pre">--nobind</span></tt> option.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--mark <var>value</var></span></kbd></td>
<td>Mark encrypted packets being sent with <tt class="docutils literal">value</tt>. The mark value can be
matched in policy routing and packetfilter rules. This option is only
supported in Linux and does nothing on other operating systems.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--mode <var>m</var></span></kbd></td>
<td>Set OpenVPN major mode. By default, OpenVPN runs in point-to-point mode
(<code>p2p</code>). OpenVPN 2.0 introduces a new mode (<code>server</code>) which
implements a multi-client server capability.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--mssfix <var>args</var></span></kbd></td>
<td><p class="first">Valid syntax:</p>
<pre class="literal-block">
mssfix max [mtu]
mssfix max [fixed]
mssfix
</pre>
<p>Announce to TCP sessions running over the tunnel that they should limit
their send packet sizes such that after OpenVPN has encapsulated them,
the resulting UDP packet size that OpenVPN sends to its peer will not
exceed <tt class="docutils literal">max</tt> bytes. The default value is <code>1492 mtu</code>. Use <code>0</code>
as max to disable mssfix.</p>
<p>If the <code>mtu</code> parameter is specified the <tt class="docutils literal">max</tt> value is interpreted
as the resulting packet size of VPN packets including the IP and UDP header.
Support for the <code>mtu</code> parameter was added with OpenVPN version 2.6.0.</p>
<p>If the <code>mtu</code> parameter is not specified, the <tt class="docutils literal">max</tt> parameter
is interpreted in the same way as the <tt class="docutils literal"><span class="pre">--link-mtu</span></tt> parameter, i.e.
the UDP packet size after encapsulation overhead has been added in, but
not including the UDP header itself. Resulting packet would be at most 28
bytes larger for IPv4 and 48 bytes for IPv6 (20/40 bytes for IP header and
8 bytes for UDP header). Default value of 1450 allows OpenVPN packets to be
transmitted over IPv4 on a link with MTU 1478 or higher without IP level
fragmentation (and 1498 for IPv6).</p>
<p>If the <code>fixed</code> parameter is specified, OpenVPN will make no attempt
to calculate the VPN encapsulation overhead but instead will set the MSS to
limit the size of the payload IP packets to the specified number. IPv4 packets
will have the MSS value lowered to mssfix - 40 and IPv6 packets to mssfix - 60.</p>
<p>if <tt class="docutils literal"><span class="pre">--mssfix</span></tt> is specified is specified without any parameter it
inherits the parameters of <tt class="docutils literal"><span class="pre">--fragment</span></tt> if specified or uses the
default for <tt class="docutils literal"><span class="pre">--mssfix</span></tt> otherwise.</p>
<p>The <tt class="docutils literal"><span class="pre">--mssfix</span></tt> option only makes sense when you are using the UDP
protocol for OpenVPN peer-to-peer communication, i.e. <tt class="docutils literal"><span class="pre">--proto</span> udp</tt>.</p>
<p><tt class="docutils literal"><span class="pre">--mssfix</span></tt> and <tt class="docutils literal"><span class="pre">--fragment</span></tt> can be ideally used together, where
<tt class="docutils literal"><span class="pre">--mssfix</span></tt> will try to keep TCP from needing packet fragmentation in
the first place, and if big packets come through anyhow (from protocols
other than TCP), <tt class="docutils literal"><span class="pre">--fragment</span></tt> will internally fragment them.</p>
<p><tt class="docutils literal"><span class="pre">--max-packet-size</span></tt>, <tt class="docutils literal"><span class="pre">--fragment</span></tt>, and <tt class="docutils literal"><span class="pre">--mssfix</span></tt> are designed to
work around cases where Path MTU discovery is broken on the network path
between OpenVPN peers.</p>
<p>The usual symptom of such a breakdown is an OpenVPN connection which
successfully starts, but then stalls during active usage.</p>
<p>If <tt class="docutils literal"><span class="pre">--fragment</span></tt> and <tt class="docutils literal"><span class="pre">--mssfix</span></tt> are used together, <tt class="docutils literal"><span class="pre">--mssfix</span></tt> will
take its default <tt class="docutils literal">max</tt> parameter from the <tt class="docutils literal"><span class="pre">--fragment</span> max</tt> option.</p>
<p>Therefore, one could lower the maximum UDP packet size to 1300 (a good
first try for solving MTU-related connection problems) with the
following options:</p>
<pre class="literal-block">
--tun-mtu 1500 --fragment 1300 --mssfix
</pre>
<p class="last">If the <tt class="docutils literal"><span class="pre">max-packet-size</span> size</tt> option is used in the configuration
it will also act as if <tt class="docutils literal">mssfix size mtu</tt> was specified in the
configuration.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--mtu-disc <var>type</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Should we do Path MTU discovery on TCP/UDP channel? Only supported on
OSes such as Linux that supports the necessary system call to set.</p>
<p>Valid types:</p>
<p><code>no</code> Never send DF (Don't Fragment) frames</p>
<p><code>maybe</code> Use per-route hints</p>
<p class="last"><code>yes</code> Always DF (Don't Fragment)</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--mtu-test</span></kbd></td>
<td>To empirically measure MTU on connection startup, add the <tt class="docutils literal"><span class="pre">--mtu-test</span></tt>
option to your configuration. OpenVPN will send ping packets of various
sizes to the remote peer and measure the largest packets which were
successfully received. The <tt class="docutils literal"><span class="pre">--mtu-test</span></tt> process normally takes about 3
minutes to complete.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--nobind</span></kbd></td>
<td>Do not bind to local address and port. The IP stack will allocate a
dynamic port for returning packets. Since the value of the dynamic port
could not be known in advance by a peer, this option is only suitable
for peers which will be initiating connections by using the <tt class="docutils literal"><span class="pre">--remote</span></tt>
option.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--passtos</span></kbd></td>
<td>Set the TOS field of the tunnel packet to what the payload's TOS is.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--ping <var>n</var></span></kbd></td>
<td><p class="first">Ping remote over the TCP/UDP control channel if no packets have been
sent for at least <tt class="docutils literal">n</tt> seconds (specify <tt class="docutils literal"><span class="pre">--ping</span></tt> on both peers to
cause ping packets to be sent in both directions since OpenVPN ping
packets are not echoed like IP ping packets). When used in one of
OpenVPN's secure modes (where <tt class="docutils literal"><span class="pre">--secret</span></tt>, <tt class="docutils literal"><span class="pre">--tls-server</span></tt> or
<tt class="docutils literal"><span class="pre">--tls-client</span></tt> is specified), the ping packet will be
cryptographically secure.</p>
<p>This option has two intended uses:</p>
<ol class="arabic simple">
<li>Compatibility with stateful firewalls. The periodic ping will ensure
that a stateful firewall rule which allows OpenVPN UDP packets to
pass will not time out.</li>
<li>To provide a basis for the remote to test the existence of its peer
using the <tt class="docutils literal"><span class="pre">--ping-exit</span></tt> option.</li>
</ol>
<p class="last">When using OpenVPN in server mode see also <tt class="docutils literal"><span class="pre">--keepalive</span></tt>.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--ping-exit <var>n</var></span></kbd></td>
<td><p class="first">Causes OpenVPN to exit after <tt class="docutils literal">n</tt> seconds pass without reception of a
ping or other packet from remote. This option can be combined with
<tt class="docutils literal"><span class="pre">--inactive</span></tt>, <tt class="docutils literal"><span class="pre">--ping</span></tt> and <tt class="docutils literal"><span class="pre">--ping-exit</span></tt> to create a two-tiered
inactivity disconnect.</p>
<p>For example,</p>
<pre class="literal-block">
openvpn [options...] --inactive 3600 --ping 10 --ping-exit 60
</pre>
<p class="last">when used on both peers will cause OpenVPN to exit within 60 seconds if
its peer disconnects, but will exit after one hour if no actual tunnel
data is exchanged.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ping-restart <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Similar to <tt class="docutils literal"><span class="pre">--ping-exit</span></tt>, but trigger a <code>SIGUSR1</code> restart after
<tt class="docutils literal">n</tt> seconds pass without reception of a ping or other packet from
remote.</p>
<p>This option is useful in cases where the remote peer has a dynamic IP
address and a low-TTL DNS name is used to track the IP address using a
service such as <a class="reference external" href="https://www.nsupdate.info/">https://www.nsupdate.info/</a> + a dynamic DNS client such as
<tt class="docutils literal">ddclient</tt>.</p>
<p>If the peer cannot be reached, a restart will be triggered, causing the
hostname used with <tt class="docutils literal"><span class="pre">--remote</span></tt> to be re-resolved (if <tt class="docutils literal"><span class="pre">--resolv-retry</span></tt>
is also specified).</p>
<p>In server mode, <tt class="docutils literal"><span class="pre">--ping-restart</span></tt>, <tt class="docutils literal"><span class="pre">--inactive</span></tt> or any other type of
internally generated signal will always be applied to individual client
instance objects, never to whole server itself. Note also in server mode
that any internally generated signal which would normally cause a
restart, will cause the deletion of the client instance object instead.</p>
<p>In client mode, the <tt class="docutils literal"><span class="pre">--ping-restart</span></tt> parameter is set to 120 seconds
by default. This default will hold until the client pulls a replacement
value from the server, based on the <tt class="docutils literal"><span class="pre">--keepalive</span></tt> setting in the
server configuration. To disable the 120 second default, set
<tt class="docutils literal"><span class="pre">--ping-restart</span> 0</tt> on the client.</p>
<p>See the signals section below for more information on <code>SIGUSR1</code>.</p>
<p>Note that the behavior of <tt class="docutils literal">SIGUSR1</tt> can be modified by the
<tt class="docutils literal"><span class="pre">--persist-tun</span></tt>, <tt class="docutils literal"><span class="pre">--persist-key</span></tt>, <tt class="docutils literal"><span class="pre">--persist-local-ip</span></tt> and
<tt class="docutils literal"><span class="pre">--persist-remote-ip</span></tt> options.</p>
<p class="last">Also note that <tt class="docutils literal"><span class="pre">--ping-exit</span></tt> and <tt class="docutils literal"><span class="pre">--ping-restart</span></tt> are mutually
exclusive and cannot be used together.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ping-timer-rem</span></kbd></td>
</tr>
<tr><td> </td><td>Run the <tt class="docutils literal"><span class="pre">--ping-exit</span></tt> / <tt class="docutils literal"><span class="pre">--ping-restart</span></tt> timer only if we have a
remote address. Use this option if you are starting the daemon in listen
mode (i.e. without an explicit <tt class="docutils literal"><span class="pre">--remote</span></tt> peer), and you don't want to
start clocking timeouts until a remote peer connects.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--proto <var>p</var></span></kbd></td>
<td><p class="first">Use protocol <tt class="docutils literal">p</tt> for communicating with remote host. <tt class="docutils literal">p</tt> can be
<code>udp</code>, <code>tcp-client</code>, or <code>tcp-server</code>. You can also
limit OpenVPN to use only IPv4 or only IPv6 by specifying <tt class="docutils literal">p</tt> as
<code>udp4</code>, <code>tcp4-client</code>, <code>tcp4-server</code> or <code>udp6</code>,
<code>tcp6-client</code>, <code>tcp6-server</code>, respectively.</p>
<p>The default protocol is <code>udp</code> when <tt class="docutils literal"><span class="pre">--proto</span></tt> is not specified.</p>
<p>For UDP operation, <tt class="docutils literal"><span class="pre">--proto</span> udp</tt> should be specified on both peers.</p>
<p>For TCP operation, one peer must use <tt class="docutils literal"><span class="pre">--proto</span> <span class="pre">tcp-server</span></tt> and the
other must use <tt class="docutils literal"><span class="pre">--proto</span> <span class="pre">tcp-client</span></tt>. A peer started with
<code>tcp-server</code> will wait indefinitely for an incoming connection. A peer
started with <code>tcp-client</code> will attempt to connect, and if that fails,
will sleep for 5 seconds (adjustable via the <tt class="docutils literal"><span class="pre">--connect-retry</span></tt> option)
and try again infinite or up to N retries (adjustable via the
<tt class="docutils literal"><span class="pre">--connect-retry-max</span></tt> option). Both TCP client and server will
simulate a SIGUSR1 restart signal if either side resets the connection.</p>
<p>OpenVPN is designed to operate optimally over UDP, but TCP capability is
provided for situations where UDP cannot be used. In comparison with
UDP, TCP will usually be somewhat less efficient and less robust when
used over unreliable or congested networks.</p>
<p>This article outlines some of problems with tunneling IP over TCP:
<a class="reference external" href="http://sites.inka.de/sites/bigred/devel/tcp-tcp.html">http://sites.inka.de/sites/bigred/devel/tcp-tcp.html</a></p>
<p class="last">There are certain cases, however, where using TCP may be advantageous
from a security and robustness perspective, such as tunneling non-IP or
application-level UDP protocols, or tunneling protocols which don't
possess a built-in reliability layer.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--port <var>port</var></span></kbd></td>
<td>TCP/UDP port number or port name for both local and remote (sets both
<tt class="docutils literal"><span class="pre">--lport</span></tt> and <tt class="docutils literal"><span class="pre">--rport</span></tt> options to given port). The current default
of 1194 represents the official IANA port number assignment for OpenVPN
and has been used since version 2.0-beta17. Previous versions used port
5000 as the default.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--rport <var>port</var></span></kbd></td>
<td>Set TCP/UDP port number or name used by the <tt class="docutils literal"><span class="pre">--remote</span></tt> option. The
port can also be set directly using the <tt class="docutils literal"><span class="pre">--remote</span></tt> option.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--replay-window <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Modify the replay protection sliding-window size and time window.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
replay-window n
replay-window n t
</pre>
<p>Use a replay protection sliding-window of size <tt class="docutils literal">n</tt> and a time window
of <tt class="docutils literal">t</tt> seconds.</p>
<p>By default <tt class="docutils literal">n</tt> is <code>64</code> (the IPSec default) and <tt class="docutils literal">t</tt> is
<code>15</code> seconds.</p>
<p>This option is only relevant in UDP mode, i.e. when either <tt class="docutils literal"><span class="pre">--proto</span>
udp</tt> is specified, or no <tt class="docutils literal"><span class="pre">--proto</span></tt> option is specified.</p>
<p>When OpenVPN tunnels IP packets over UDP, there is the possibility that
packets might be dropped or delivered out of order. Because OpenVPN,
like IPSec, is emulating the physical network layer, it will accept an
out-of-order packet sequence, and will deliver such packets in the same
order they were received to the TCP/IP protocol stack, provided they
satisfy several constraints.</p>
<ol class="loweralpha simple">
<li>The packet cannot be a replay (unless <tt class="docutils literal"><span class="pre">--no-replay</span></tt> is
specified, which disables replay protection altogether).</li>
<li>If a packet arrives out of order, it will only be accepted if
the difference between its sequence number and the highest sequence
number received so far is less than <tt class="docutils literal">n</tt>.</li>
<li>If a packet arrives out of order, it will only be accepted if it
arrives no later than <tt class="docutils literal">t</tt> seconds after any packet containing a higher
sequence number.</li>
</ol>
<p>If you are using a network link with a large pipeline (meaning that the
product of bandwidth and latency is high), you may want to use a larger
value for <tt class="docutils literal">n</tt>. Satellite links in particular often require this.</p>
<p>If you run OpenVPN at <tt class="docutils literal"><span class="pre">--verb</span> 4</tt>, you will see the message
"PID_ERR replay-window backtrack occurred [x]" every time the maximum sequence
number backtrack seen thus far increases. This can be used to calibrate
<tt class="docutils literal">n</tt>.</p>
<p>There is some controversy on the appropriate method of handling packet
reordering at the security layer.</p>
<p>Namely, to what extent should the security layer protect the
encapsulated protocol from attacks which masquerade as the kinds of
normal packet loss and reordering that occur over IP networks?</p>
<p>The IPSec and OpenVPN approach is to allow packet reordering within a
certain fixed sequence number window.</p>
<p>OpenVPN adds to the IPSec model by limiting the window size in time as
well as sequence space.</p>
<p>OpenVPN also adds TCP transport as an option (not offered by IPSec) in
which case OpenVPN can adopt a very strict attitude towards message
deletion and reordering: Don't allow it. Since TCP guarantees
reliability, any packet loss or reordering event can be assumed to be an
attack.</p>
<p>In this sense, it could be argued that TCP tunnel transport is preferred
when tunneling non-IP or UDP application protocols which might be
vulnerable to a message deletion or reordering attack which falls within
the normal operational parameters of IP networks.</p>
<p class="last">So I would make the statement that one should never tunnel a non-IP
protocol or UDP application protocol over UDP, if the protocol might be
vulnerable to a message deletion or reordering attack that falls within
the normal operating parameters of what is to be expected from the
physical IP layer. The problem is easily fixed by simply using TCP as
the VPN transport layer.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--replay-persist <var>file</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Persist replay-protection state across sessions using <tt class="docutils literal">file</tt> to save
and reload the state.</p>
<p>This option will keep a disk copy of the current replay protection state
(i.e. the most recent packet timestamp and sequence number received from
the remote peer), so that if an OpenVPN session is stopped and
restarted, it will reject any replays of packets which were already
received by the prior session.</p>
<p class="last">This option only makes sense when replay protection is enabled (the
default) and you are using either <tt class="docutils literal"><span class="pre">--secret</span></tt> (shared-secret key mode)
or TLS mode with <tt class="docutils literal"><span class="pre">--tls-auth</span></tt>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--session-timeout <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Raises <code>SIGTERM</code> for the client instance after <tt class="docutils literal">n</tt> seconds since
the beginning of the session, forcing OpenVPN to disconnect.
In client mode, OpenVPN will disconnect and exit, while in server mode
all client sessions are terminated.</p>
<p class="last">This option can also be specified in a client instance config file
using <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> or dynamically generated using a
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> script. In these cases, only the related client
session is terminated.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--socket-flags <var>flags</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Apply the given flags to the OpenVPN transport socket. Currently, only
<code>TCP_NODELAY</code> is supported.</p>
<p>The <code>TCP_NODELAY</code> socket flag is useful in TCP mode, and causes the
kernel to send tunnel packets immediately over the TCP connection without
trying to group several smaller packets into a larger packet. This can
result in a considerably improvement in latency.</p>
<p class="last">This option is pushable from server to client, and should be used on
both client and server for maximum effect.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--tcp-nodelay</span></kbd></td>
<td><p class="first">This macro sets the <code>TCP_NODELAY</code> socket flag on the server as well
as pushes it to connecting clients. The <code>TCP_NODELAY</code> flag disables
the Nagle algorithm on TCP sockets causing packets to be transmitted
immediately with low latency, rather than waiting a short period of time
in order to aggregate several packets into a larger containing packet.
In VPN applications over TCP, <code>TCP_NODELAY</code> is generally a good
latency optimization.</p>
<p>The macro expands as follows:</p>
<pre class="last literal-block">
if mode server:
socket-flags TCP_NODELAY
push "socket-flags TCP_NODELAY"
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--max-packet-size <var>size</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">This option will instruct OpenVPN to try to limit the maximum on-write packet
size by restricting the control channel packet size and setting <tt class="docutils literal"><span class="pre">--mssfix</span></tt>.</p>
<p>OpenVPN will try to keep its control channel messages below this size but
due to some constraints in the protocol this is not always possible. If the
option is not set, the control packet maximum size defaults to 1250.
The control channel packet size will be restricted to values between
154 and 2048. The maximum packet size includes encapsulation overhead like
UDP and IP.</p>
<p>In terms of <tt class="docutils literal"><span class="pre">--mssfix</span></tt> it will expand to:</p>
<pre class="literal-block">
mssfix size mtu
</pre>
<p>If you need to set <tt class="docutils literal"><span class="pre">--mssfix</span></tt> for data channel and control channel maximum
packet size independently, use <tt class="docutils literal"><span class="pre">--max-packet-size</span></tt> first, followed by a
<tt class="docutils literal"><span class="pre">--mssfix</span></tt> in the configuration.</p>
<p class="last">In general the default size of 1250 should work almost universally apart
from specific corner cases, especially since IPv6 requires a MTU of 1280
or larger.</p>
</td></tr>
</tbody>
</table>
</div>
<div class="section" id="virtual-network-adapter-vpn-interface">
<h2>Virtual Network Adapter (VPN interface)</h2>
<p>Options in this section relates to configuration of the virtual tun/tap
network interface, including setting the VPN IP address and network
routing.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--bind-dev <var>device</var></span></kbd></td>
</tr>
<tr><td> </td><td>(Linux only) Set <tt class="docutils literal">device</tt> to bind the server socket to a
<a class="reference internal" href="#virtual-routing-and-forwarding">Virtual Routing and Forwarding</a> device</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--block-ipv6</span></kbd></td>
<td><p class="first">On the client, instead of sending IPv6 packets over the VPN tunnel, all
IPv6 packets are answered with an ICMPv6 no route host message. On the
server, all IPv6 packets from clients are answered with an ICMPv6 no
route to host message. This options is intended for cases when IPv6
should be blocked and other options are not available. <tt class="docutils literal"><span class="pre">--block-ipv6</span></tt>
will use the remote IPv6 as source address of the ICMPv6 packets if set,
otherwise will use <code>fe80::7</code> as source address.</p>
<p>For this option to make sense you actually have to route traffic to the
tun interface. The following example config block would send all IPv6
traffic to OpenVPN and answer all requests with no route to host,
effectively blocking IPv6 (to avoid IPv6 connections from dual-stacked
clients leaking around IPv4-only VPN services).</p>
<dl class="docutils">
<dt><strong>Client config</strong></dt>
<dd><pre class="first last literal-block">
--ifconfig-ipv6 fd15:53b6:dead::2/64 fd15:53b6:dead::1
--redirect-gateway ipv6
--block-ipv6
</pre>
</dd>
<dt><strong>Server config</strong></dt>
<dd><p class="first">Push a "valid" ipv6 config to the client and block on the server</p>
<pre class="last literal-block">
--push "ifconfig-ipv6 fd15:53b6:dead::2/64 fd15:53b6:dead::1"
--push "redirect-gateway ipv6"
--block-ipv6
</pre>
</dd>
</dl>
<p class="last">Note: this option does not influence traffic sent from the server
towards the client (neither on the server nor on the client side).
This is not seen as necessary, as such traffic can be most easily
avoided by not configuring IPv6 on the server tun, or setting up a
server-side firewall rule.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--dev <var>device</var></span></kbd></td>
<td><p class="first">TUN/TAP virtual network device which can be <code>tunX</code>, <code>tapX</code>,
<code>null</code> or an arbitrary name string (<code>X</code> can be omitted for
a dynamic device.)</p>
<p>See examples section below for an example on setting up a TUN device.</p>
<p>You must use either tun devices on both ends of the connection or tap
devices on both ends. You cannot mix them, as they represent different
underlying network layers:</p>
<dl class="docutils">
<dt><code>tun</code></dt>
<dd>devices encapsulate IPv4 or IPv6 (OSI Layer 3)</dd>
<dt><code>tap</code></dt>
<dd>devices encapsulate Ethernet 802.3 (OSI Layer 2).</dd>
</dl>
<p>Valid syntaxes:</p>
<pre class="literal-block">
dev tun2
dev tap4
dev ovpn
</pre>
<p>What happens if the device name is not <code>tun</code> or <code>tap</code> is
platform dependent.</p>
<p>On most platforms, <code>tunN</code> (e.g. tun2, tun30) and <code>tapN</code>
(e.g. tap3) will create a numbered tun/tap interface with the number
specified - this is useful if multiple OpenVPN instances are active,
and the instance-to-device mapping needs to be known. Some platforms
do not support "numbered tap", so trying <tt class="docutils literal"><span class="pre">--dev</span> tap3</tt> will fail.</p>
<p>Arbitrary device names (e.g. <tt class="docutils literal"><span class="pre">--dev</span> <span class="pre">tun-home</span></tt>) will only work on
FreeBSD (with the DCO kernel driver for <tt class="docutils literal">tun</tt> devices) and Linux
(for both <tt class="docutils literal">tun</tt> and <tt class="docutils literal">tap</tt> devices, DCO and tun/tap driver).</p>
<p>If such a device name starts with <tt class="docutils literal">tun</tt> or <tt class="docutils literal">tap</tt> (e.g. <tt class="docutils literal"><span class="pre">tun-home</span></tt>),
OpenVPN will choose the right device type automatically. Otherwise the
desired device type needs to be specified with <tt class="docutils literal"><span class="pre">--dev-type</span> tun</tt> or
<tt class="docutils literal"><span class="pre">--dev-type</span> tap</tt>.</p>
<p class="last">On Windows, only the names <code>tun</code> and <code>tap</code> are supported.
Selection among multiple installed drivers or driver instances is done
with <tt class="docutils literal"><span class="pre">--dev-node</span></tt> and <tt class="docutils literal"><span class="pre">--windows-driver</span></tt>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--dev-node <var>node</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">This is a highly system dependent option to influence tun/tap driver
selection.</p>
<p>On Linux, tun/tap devices are created by accessing <code>/dev/net/tun</code>,
and this device name can be changed using <tt class="docutils literal"><span class="pre">--dev-node</span> ...</tt>.</p>
<p>Under Mac OS X this option can be used to specify the default tun
implementation. Using <tt class="docutils literal"><span class="pre">--dev-node</span> utun</tt> forces usage of the native
Darwin tun kernel support. Use <tt class="docutils literal"><span class="pre">--dev-node</span> utunN</tt> to select a specific
utun instance. To force using the <code>tun.kext</code> (<code>/dev/tunX</code>)
use <tt class="docutils literal"><span class="pre">--dev-node</span> tun</tt>. When not specifying a <tt class="docutils literal"><span class="pre">--dev-node</span></tt> option
openvpn will first try to open utun, and fall back to tun.kext.</p>
<p>On Windows systems, select the TAP-Win32 adapter which is named <tt class="docutils literal">node</tt>
in the Network Connections Control Panel or the raw GUID of the adapter
enclosed by braces. The <tt class="docutils literal"><span class="pre">--show-adapters</span></tt> option under Windows can
also be used to enumerate all available TAP-Win32 adapters and will show
both the network connections control panel name and the GUID for each
TAP-Win32 adapter.</p>
<p class="last">On other platforms, <tt class="docutils literal"><span class="pre">--dev-node</span> node</tt> will influence the naming of the
created tun/tap device, if supported on that platform. If OpenVPN cannot
figure out whether <tt class="docutils literal">node</tt> is a TUN or TAP device based on the name,
you should also specify <tt class="docutils literal"><span class="pre">--dev-type</span> tun</tt> or <tt class="docutils literal"><span class="pre">--dev-type</span> tap</tt>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--dev-type <var>device-type</var></span></kbd></td>
</tr>
<tr><td> </td><td>Which device type are we using? <tt class="docutils literal"><span class="pre">device-type</span></tt> should be <code>tun</code>
(OSI Layer 3) or <code>tap</code> (OSI Layer 2). Use this option only if
the TUN/TAP device used with <tt class="docutils literal"><span class="pre">--dev</span></tt> does not begin with <code>tun</code>
or <code>tap</code>.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--dhcp-option <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Set additional network parameters on supported platforms. May be specified
on the client or pushed from the server. On Windows these options are
handled by the <tt class="docutils literal"><span class="pre">tap-windows6</span></tt> driver by default or directly by OpenVPN
if dhcp is disabled or the <tt class="docutils literal">wintun</tt> driver is in use. The
<tt class="docutils literal">OpenVPN for Android</tt> client also handles them internally.</p>
<p>On all other platforms these options are only saved in the client's
environment under the name <code>foreign_option_{n}</code> before the
<tt class="docutils literal"><span class="pre">--up</span></tt> script is called. A plugin or an <tt class="docutils literal"><span class="pre">--up</span></tt> script must be used to
pick up and interpret these as required. Many Linux distributions include
such scripts and some third-party user interfaces such as tunnelblick also
come with scripts that process these options.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
dhcp-option type [parm]
</pre>
<dl class="docutils">
<dt><code>DOMAIN</code> <tt class="docutils literal">name</tt></dt>
<dd>Set Connection-specific DNS Suffix to <code>name</code>.</dd>
<dt><code>ADAPTER_DOMAIN_SUFFIX</code> <tt class="docutils literal">name</tt></dt>
<dd>Alias to <code>DOMAIN</code>. This is a compatibility option, it
should not be used in new deployments.</dd>
<dt><code>DOMAIN-SEARCH</code> <tt class="docutils literal">name</tt></dt>
<dd>Add <code>name</code> to the domain search list.
Repeat this option to add more entries. Up to
10 domains are supported.</dd>
<dt><code>DNS</code> <tt class="docutils literal">address</tt></dt>
<dd><p class="first">Set primary domain name server IPv4 or IPv6 address.
Repeat this option to set secondary DNS server addresses.</p>
<p class="last">Note: DNS IPv6 servers are currently set using netsh (the existing
DHCP code can only do IPv4 DHCP, and that protocol only permits
IPv4 addresses anywhere). The option will be put into the
environment, so an <tt class="docutils literal"><span class="pre">--up</span></tt> script could act upon it if needed.</p>
</dd>
<dt><code>WINS</code> <tt class="docutils literal">address</tt></dt>
<dd>Set primary WINS server address (NetBIOS over TCP/IP Name Server).
Repeat this option to set secondary WINS server addresses.</dd>
<dt><code>NBDD</code> <tt class="docutils literal">address</tt></dt>
<dd>Set primary NBDD server address (NetBIOS over TCP/IP Datagram
Distribution Server). Repeat this option to set secondary NBDD
server addresses.</dd>
<dt><code>NTP</code> <tt class="docutils literal">address</tt></dt>
<dd>Set primary NTP server address (Network Time Protocol).
Repeat this option to set secondary NTP server addresses.</dd>
<dt><code>NBT</code> <tt class="docutils literal">type</tt></dt>
<dd><p class="first">Set NetBIOS over TCP/IP Node type. Possible options:</p>
<dl class="last docutils">
<dt><code>1</code></dt>
<dd>b-node (broadcasts)</dd>
<dt><code>2</code></dt>
<dd>p-node (point-to-point name queries to a WINS server)</dd>
<dt><code>4</code></dt>
<dd>m-node (broadcast then query name server)</dd>
<dt><code>8</code></dt>
<dd>h-node (query name server, then broadcast).</dd>
</dl>
</dd>
<dt><code>NBS</code> <tt class="docutils literal"><span class="pre">scope-id</span></tt></dt>
<dd>Set NetBIOS over TCP/IP Scope. A NetBIOS Scope ID provides an
extended naming service for the NetBIOS over TCP/IP (Known as NBT)
module. The primary purpose of a NetBIOS scope ID is to isolate
NetBIOS traffic on a single network to only those nodes with the
same NetBIOS scope ID. The NetBIOS scope ID is a character string
that is appended to the NetBIOS name. The NetBIOS scope ID on two
hosts must match, or the two hosts will not be able to communicate.
The NetBIOS Scope ID also allows computers to use the same computer
name, as they have different scope IDs. The Scope ID becomes a part
of the NetBIOS name, making the name unique. (This description of
NetBIOS scopes courtesy of <a class="reference external" href="mailto:NeonSurge@abyss.com">NeonSurge@abyss.com</a>)</dd>
<dt><code>DISABLE-NBT</code></dt>
<dd>Disable Netbios-over-TCP/IP.</dd>
</dl>
<table class="last docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field"><th class="field-name">code:</th><td class="field-body"><p class="first"><cite>PROXY_HTTP</cite> <tt class="docutils literal">host</tt> <tt class="docutils literal">port</tt>
Sets a HTTP proxy that should be used when connected to the VPN.</p>
<p class="last">This option currently only works on OpenVPN for Android and requires
Android 10 or later.</p>
</td>
</tr>
</tbody>
</table>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ifconfig <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Set TUN/TAP adapter parameters. It requires the <em>IP address</em> of the local
VPN endpoint. For TUN devices in point-to-point mode, the next argument
must be the VPN IP address of the remote VPN endpoint. For TAP devices,
or TUN devices used with <tt class="docutils literal"><span class="pre">--topology</span> subnet</tt>, the second argument
is the subnet mask of the virtual network segment which is being created
or connected to.</p>
<p>For TUN devices, which facilitate virtual point-to-point IP connections
(when used in <tt class="docutils literal"><span class="pre">--topology</span> net30</tt> or <tt class="docutils literal">p2p</tt> mode), the proper usage of
<tt class="docutils literal"><span class="pre">--ifconfig</span></tt> is to use two private IP addresses which are not a member
of any existing subnet which is in use. The IP addresses may be
consecutive and should have their order reversed on the remote peer.
After the VPN is established, by pinging <tt class="docutils literal">rn</tt>, you will be pinging
across the VPN.</p>
<p>For TAP devices, which provide the ability to create virtual ethernet
segments, or TUN devices in <tt class="docutils literal"><span class="pre">--topology</span> subnet</tt> mode (which create
virtual "multipoint networks"), <tt class="docutils literal"><span class="pre">--ifconfig</span></tt> is used to set an IP
address and subnet mask just as a physical ethernet adapter would be
similarly configured. If you are attempting to connect to a remote
ethernet bridge, the IP address and subnet should be set to values which
would be valid on the the bridged ethernet segment (note also that DHCP
can be used for the same purpose).</p>
<p>This option, while primarily a proxy for the <tt class="docutils literal">ifconfig</tt>(8) command,
is designed to simplify TUN/TAP tunnel configuration by providing a
standard interface to the different ifconfig implementations on
different platforms.</p>
<p><tt class="docutils literal"><span class="pre">--ifconfig</span></tt> parameters which are IP addresses can also be specified
as a DNS or /etc/hosts file resolvable name.</p>
<p>For TAP devices, <tt class="docutils literal"><span class="pre">--ifconfig</span></tt> should not be used if the TAP interface
will be getting an IP address lease from a DHCP server.</p>
<p>Examples:</p>
<pre class="last literal-block">
# tun device in net30/p2p mode
ifconfig 10.8.0.2 10.8.0.1
# tun/tap device in subnet mode
ifconfig 10.8.0.2 255.255.255.0
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ifconfig-ipv6 <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Configure an IPv6 address on the <em>tun</em> device.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
ifconfig-ipv6 ipv6addr/bits [ipv6remote]
</pre>
<p>The <tt class="docutils literal">ipv6addr/bits</tt> argument is the IPv6 address to use. The
second parameter is used as route target for <tt class="docutils literal"><span class="pre">--route-ipv6</span></tt> if no
gateway is specified.</p>
<p class="last">The <tt class="docutils literal"><span class="pre">--topology</span></tt> option has no influence with <tt class="docutils literal"><span class="pre">--ifconfig-ipv6</span></tt></p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ifconfig-noexec</span></kbd></td>
</tr>
<tr><td> </td><td>Don't actually execute ifconfig/netsh commands, instead pass
<tt class="docutils literal"><span class="pre">--ifconfig</span></tt> parameters to scripts using environmental variables.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ifconfig-nowarn</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Don't output an options consistency check warning if the <tt class="docutils literal"><span class="pre">--ifconfig</span></tt>
option on this side of the connection doesn't match the remote side.
This is useful when you want to retain the overall benefits of the
options consistency check (also see <tt class="docutils literal"><span class="pre">--disable-occ</span></tt> option) while only
disabling the ifconfig component of the check.</p>
<p>For example, if you have a configuration where the local host uses
<tt class="docutils literal"><span class="pre">--ifconfig</span></tt> but the remote host does not, use <tt class="docutils literal"><span class="pre">--ifconfig-nowarn</span></tt>
on the local host.</p>
<p class="last">This option will also silence warnings about potential address conflicts
which occasionally annoy more experienced users by triggering "false
positive" warnings.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--lladdr <var>address</var></span></kbd></td>
</tr>
<tr><td> </td><td>Specify the link layer address, more commonly known as the MAC address.
Only applied to TAP devices.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--persist-tun</span></kbd></td>
<td><p class="first">Don't close and reopen TUN/TAP device or run up/down scripts across
<code>SIGUSR1</code> or <tt class="docutils literal"><span class="pre">--ping-restart</span></tt> restarts.</p>
<p class="last"><code>SIGUSR1</code> is a restart signal similar to <code>SIGHUP</code>, but which
offers finer-grained control over reset options.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--redirect-gateway <var>flags</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Automatically execute routing commands to cause all outgoing IP traffic
to be redirected over the VPN. This is a client-side option.</p>
<p>This option performs three steps:</p>
<ol class="arabic simple">
<li>Create a static route for the <tt class="docutils literal"><span class="pre">--remote</span></tt> address which
forwards to the pre-existing default gateway. This is done so that
<tt class="docutils literal">(3)</tt> will not create a routing loop.</li>
<li>Delete the default gateway route.</li>
<li>Set the new default gateway to be the VPN endpoint address
(derived either from <tt class="docutils literal"><span class="pre">--route-gateway</span></tt> or the second parameter to
<tt class="docutils literal"><span class="pre">--ifconfig</span></tt> when <tt class="docutils literal"><span class="pre">--dev</span> tun</tt> is specified).</li>
</ol>
<p>When the tunnel is torn down, all of the above steps are reversed so
that the original default route is restored.</p>
<p>Option flags:</p>
<dl class="last docutils">
<dt><code>local</code></dt>
<dd>Add the <code>local</code> flag if both OpenVPN peers are directly
connected via a common subnet, such as with wireless. The
<code>local</code> flag will cause step <tt class="docutils literal">(1)</tt> above to be omitted.</dd>
<dt><code>autolocal</code></dt>
<dd>Try to automatically determine whether to enable <code>local</code>
flag above.</dd>
<dt><code>def1</code></dt>
<dd>Use this flag to override the default gateway by using
<code>0.0.0.0/1</code> and <code>128.0.0.0/1</code> rather than
<code>0.0.0.0/0</code>. This has the benefit of overriding but not
wiping out the original default gateway.</dd>
<dt><code>bypass-dhcp</code></dt>
<dd>Add a direct route to the DHCP server (if it is non-local) which
bypasses the tunnel (Available on Windows clients, may not be
available on non-Windows clients).</dd>
<dt><code>bypass-dns</code></dt>
<dd>Add a direct route to the DNS server(s) (if they are non-local)
which bypasses the tunnel (Available on Windows clients, may
not be available on non-Windows clients).</dd>
<dt><code>block-local</code></dt>
<dd>Block access to local LAN when the tunnel is active, except for
the LAN gateway itself. This is accomplished by routing the local
LAN (except for the LAN gateway address) into the tunnel.</dd>
<dt><code>ipv6</code></dt>
<dd>Redirect IPv6 routing into the tunnel. This works similar to
the <code>def1</code> flag, that is, more specific IPv6 routes are added
(<code>2000::/4</code>, <code>3000::/4</code>), covering the whole IPv6
unicast space.</dd>
<dt><code>!ipv4</code></dt>
<dd>Do not redirect IPv4 traffic - typically used in the flag pair
<code>ipv6 !ipv4</code> to redirect IPv6-only.</dd>
</dl>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--redirect-private <var>flags</var></span></kbd></td>
</tr>
<tr><td> </td><td>Like <tt class="docutils literal"><span class="pre">--redirect-gateway</span></tt>, but omit actually changing the default gateway.
Useful when pushing private subnets.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--route <var>args</var></span></kbd></td>
<td><p class="first">Add route to routing table after connection is established. Multiple
routes can be specified. Routes will be automatically torn down in
reverse order prior to TUN/TAP device close.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
route network/IP
route network/IP netmask
route network/IP netmask gateway
route network/IP netmask gateway metric
</pre>
<p>This option is intended as a convenience proxy for the <tt class="docutils literal">route</tt>(8)
shell command, while at the same time providing portable semantics
across OpenVPN's platform space.</p>
<dl class="docutils">
<dt><tt class="docutils literal">netmask</tt></dt>
<dd>defaults to <code>255.255.255.255</code> when not given</dd>
<dt><tt class="docutils literal">gateway</tt></dt>
<dd>default taken from <tt class="docutils literal"><span class="pre">--route-gateway</span></tt> or the second
parameter to <tt class="docutils literal"><span class="pre">--ifconfig</span></tt> when <tt class="docutils literal"><span class="pre">--dev</span> tun</tt> is specified.</dd>
<dt><tt class="docutils literal">metric</tt></dt>
<dd>default taken from <tt class="docutils literal"><span class="pre">--route-metric</span></tt> if set, otherwise <code>0</code>.</dd>
</dl>
<p>The default can be specified by leaving an option blank or setting it to
<code>default</code>.</p>
<p>The <tt class="docutils literal">network</tt> and <tt class="docutils literal">gateway</tt> parameters can also be specified as a
DNS or <code>/etc/hosts</code> file resolvable name, or as one of three special
keywords:</p>
<dl class="last docutils">
<dt><code>vpn_gateway</code></dt>
<dd>The remote VPN endpoint address (derived either from
<tt class="docutils literal"><span class="pre">--route-gateway</span></tt> or the second parameter to <tt class="docutils literal"><span class="pre">--ifconfig</span></tt>
when <tt class="docutils literal"><span class="pre">--dev</span> tun</tt> is specified).</dd>
<dt><code>net_gateway</code></dt>
<dd>The pre-existing IP default gateway, read from the
routing table (not supported on all OSes).</dd>
<dt><code>remote_host</code></dt>
<dd>The <tt class="docutils literal"><span class="pre">--remote</span></tt> address if OpenVPN is being run in
client mode, and is undefined in server mode.</dd>
</dl>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--route-delay <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Valid syntaxes:</p>
<pre class="literal-block">
route-delay
route-delay n
route-delay n w
</pre>
<p>Delay <tt class="docutils literal">n</tt> seconds (default <code>0</code>) after connection establishment,
before adding routes. If <tt class="docutils literal">n</tt> is <code>0</code>, routes will be added
immediately upon connection establishment. If <tt class="docutils literal"><span class="pre">--route-delay</span></tt> is
omitted, routes will be added immediately after TUN/TAP device open and
<tt class="docutils literal"><span class="pre">--up</span></tt> script execution, before any <tt class="docutils literal"><span class="pre">--user</span></tt> or <tt class="docutils literal"><span class="pre">--group</span></tt> privilege
downgrade (or <tt class="docutils literal"><span class="pre">--chroot</span></tt> execution.)</p>
<p>This option is designed to be useful in scenarios where DHCP is used to
set tap adapter addresses. The delay will give the DHCP handshake time
to complete before routes are added.</p>
<p class="last">On Windows, <tt class="docutils literal"><span class="pre">--route-delay</span></tt> tries to be more intelligent by waiting
<tt class="docutils literal">w</tt> seconds (default <code>30</code>) for the TAP-Win32 adapter
to come up before adding routes.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--route-ipv6 <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Setup IPv6 routing in the system to send the specified IPv6 network into
OpenVPN's <em>tun</em>.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
route-ipv6 ipv6addr/bits [gateway] [metric]
</pre>
<p class="last">The gateway parameter is only used for IPv6 routes across <em>tap</em> devices,
and if missing, the <tt class="docutils literal">ipv6remote</tt> field from <tt class="docutils literal"><span class="pre">--ifconfig-ipv6</span></tt> or
<tt class="docutils literal"><span class="pre">--route-ipv6-gateway</span></tt> is used.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--route-gateway <var>arg</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Specify a default <em>gateway</em> for use with <tt class="docutils literal"><span class="pre">--route</span></tt>.</p>
<p>If <code>dhcp</code> is specified as the parameter, the gateway address will
be extracted from a DHCP negotiation with the OpenVPN server-side LAN.</p>
<p>Valid syntaxes:</p>
<pre class="last literal-block">
route-gateway gateway
route-gateway dhcp
</pre>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--route-ipv6-gateway <var>gw</var></span></kbd></td>
</tr>
<tr><td> </td><td>Specify a default gateway <tt class="docutils literal">gw</tt> for use with <tt class="docutils literal"><span class="pre">--route-ipv6</span></tt>.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--route-metric <var>m</var></span></kbd></td>
</tr>
<tr><td> </td><td>Specify a default metric <tt class="docutils literal">m</tt> for use with <tt class="docutils literal"><span class="pre">--route</span></tt>.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--route-noexec</span></kbd></td>
<td>Don't add or remove routes automatically. Instead pass routes to
<tt class="docutils literal"><span class="pre">--route-up</span></tt> script using environmental variables.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--route-nopull</span></kbd></td>
<td><p class="first">When used with <tt class="docutils literal"><span class="pre">--client</span></tt> or <tt class="docutils literal"><span class="pre">--pull</span></tt>, accept options pushed by
server EXCEPT for routes, block-outside-dns and dhcp options like DNS
servers.</p>
<p class="last">When used on the client, this option effectively bars the server from
adding routes to the client's routing table, however note that this
option still allows the server to set the TCP/IP properties of the
client's TUN/TAP interface.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--topology <var>mode</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Configure virtual addressing topology when running in <tt class="docutils literal"><span class="pre">--dev</span> tun</tt>
mode. This directive has no meaning in <tt class="docutils literal"><span class="pre">--dev</span> tap</tt> mode, which always
uses a <code>subnet</code> topology.</p>
<p>If you set this directive on the server, the <tt class="docutils literal"><span class="pre">--server</span></tt> and
<tt class="docutils literal"><span class="pre">--server-bridge</span></tt> directives will automatically push your chosen
topology setting to clients as well. This directive can also be manually
pushed to clients. Like the <tt class="docutils literal"><span class="pre">--dev</span></tt> directive, this directive must
always be compatible between client and server.</p>
<p><tt class="docutils literal">mode</tt> can be one of:</p>
<dl class="docutils">
<dt><code>net30</code></dt>
<dd>Use a point-to-point topology, by allocating one /30 subnet
per client. This is designed to allow point-to-point semantics when some
or all of the connecting clients might be Windows systems. This is the
default.</dd>
<dt><code>p2p</code></dt>
<dd>Use a point-to-point topology where the remote endpoint of
the client's tun interface always points to the local endpoint of the
server's tun interface. This mode allocates a single IP address per
connecting client. Only use when none of the connecting clients are
Windows systems.</dd>
<dt><code>subnet</code></dt>
<dd>Use a subnet rather than a point-to-point topology by
configuring the tun interface with a local IP address and subnet mask,
similar to the topology used in <tt class="docutils literal"><span class="pre">--dev</span> tap</tt> and ethernet bridging
mode. This mode allocates a single IP address per connecting client and
works on Windows as well.</dd>
</dl>
<p class="last"><em>Note:</em> Using <tt class="docutils literal"><span class="pre">--topology</span> subnet</tt> changes the interpretation of the
arguments of <tt class="docutils literal"><span class="pre">--ifconfig</span></tt> to mean "address netmask", no longer "local
remote".</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--tun-mtu <var>args</var></span></kbd></td>
<td><p class="first">Valid syntaxes:</p>
<pre class="literal-block">
tun-mtu tun-mtu
tun-mtu tun-mtu occ-mtu
</pre>
<p>Take the TUN device MTU to be <tt class="docutils literal"><span class="pre">tun-mtu</span></tt> and derive the link MTU from it.
In most cases, you will probably want to leave this parameter set to
its default value.</p>
<p>The default for <code>tun-mtu</code> is 1500.</p>
<p>The OCC MTU can be used to avoid warnings about mismatched MTU from
clients. If <code>occ-mtu</code> is not specified, it will to default to the
tun-mtu.</p>
<p>The MTU (Maximum Transmission Units) is the maximum datagram size in
bytes that can be sent unfragmented over a particular network path.
OpenVPN requires that packets on the control and data channels be sent
unfragmented.</p>
<p>MTU problems often manifest themselves as connections which hang during
periods of active usage.</p>
<p>It's best to use the <tt class="docutils literal"><span class="pre">--fragment</span></tt> and/or <tt class="docutils literal"><span class="pre">--mssfix</span></tt> options to deal
with MTU sizing issues.</p>
<p class="last">Note: Depending on the platform, the operating system allows one to receive
packets larger than <tt class="docutils literal"><span class="pre">tun-mtu</span></tt> (e.g. Linux and FreeBSD) but other platforms
(like macOS) limit received packets to the same size as the MTU.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tun-max-mtu <var>maxmtu</var></span></kbd></td>
</tr>
<tr><td> </td><td>This configures the maximum MTU size that a server can push to <tt class="docutils literal">maxmtu</tt>,
by configuring the internal buffers to allow at least this packet size.
The default for <tt class="docutils literal">maxmtu</tt> is 1600. Currently, only increasing beyond 1600
is possible, and attempting to reduce max-mtu below 1600 will be ignored.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tun-mtu-extra <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td>Assume that the TUN/TAP device might return as many as <tt class="docutils literal">n</tt> bytes more
than the <tt class="docutils literal"><span class="pre">--tun-mtu</span></tt> size on read. This parameter defaults to 0, which
is sufficient for most TUN devices. TAP devices may introduce additional
overhead in excess of the MTU size, and a setting of 32 is the default
when TAP devices are used. This parameter only controls internal OpenVPN
buffer sizing, so there is no transmission overhead associated with
using a larger value.</td></tr>
</tbody>
</table>
</div>
<div class="section" id="tun-tap-standalone-operations">
<h2>TUN/TAP standalone operations</h2>
<p>These two standalone operations will require <tt class="docutils literal"><span class="pre">--dev</span></tt> and optionally
<tt class="docutils literal"><span class="pre">--user</span></tt> and/or <tt class="docutils literal"><span class="pre">--group</span></tt>.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group">
<kbd><span class="option">--mktun</span></kbd></td>
<td><p class="first">(Standalone) Create a persistent tunnel on platforms which support them
such as Linux. Normally TUN/TAP tunnels exist only for the period of
time that an application has them open. This option takes advantage of
the TUN/TAP driver's ability to build persistent tunnels that live
through multiple instantiations of OpenVPN and die only when they are
deleted or the machine is rebooted.</p>
<p>One of the advantages of persistent tunnels is that they eliminate the
need for separate <tt class="docutils literal"><span class="pre">--up</span></tt> and <tt class="docutils literal"><span class="pre">--down</span></tt> scripts to run the appropriate
<tt class="docutils literal">ifconfig</tt>(8) and <tt class="docutils literal">route</tt>(8) commands. These commands can be
placed in the the same shell script which starts or terminates an
OpenVPN session.</p>
<p>Another advantage is that open connections through the TUN/TAP-based
tunnel will not be reset if the OpenVPN peer restarts. This can be
useful to provide uninterrupted connectivity through the tunnel in the
event of a DHCP reset of the peer's public IP address (see the
<tt class="docutils literal"><span class="pre">--ipchange</span></tt> option above).</p>
<p>One disadvantage of persistent tunnels is that it is harder to
automatically configure their MTU value (see <tt class="docutils literal"><span class="pre">--link-mtu</span></tt> and
<tt class="docutils literal"><span class="pre">--tun-mtu</span></tt> above).</p>
<p class="last">On some platforms such as Windows, TAP-Win32 tunnels are persistent by
default.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--rmtun</span></kbd></td>
<td>(Standalone) Remove a persistent tunnel.</td></tr>
</tbody>
</table>
</div>
<div class="section" id="virtual-routing-and-forwarding">
<h2>Virtual Routing and Forwarding</h2>
<p>Options in this section relates to configuration of virtual routing and
forwarding in combination with the underlying operating system.</p>
<p>As of today this is only supported on Linux, a kernel >= 4.9 is
recommended.</p>
<p>This could come in handy when for example the external network should be
only used as a means to connect to some VPN endpoints and all regular
traffic should only be routed through any tunnel(s). This could be
achieved by setting up a VRF and configuring the interface connected to
the external network to be part of the VRF. The examples below will cover
this setup.</p>
<p>Another option would be to put the tun/tap interface into a VRF. This could
be done by an up-script which uses the <code>ip link set</code> command shown
below.</p>
<div class="section" id="vrf-setup-with-iproute2">
<h3>VRF setup with iproute2</h3>
<p>Create VRF <code>vrf_external</code> and map it to routing table <code>1023</code></p>
<pre class="literal-block">
ip link add vrf_external type vrf table 1023
</pre>
<p>Move <code>eth0</code> into <code>vrf_external</code></p>
<pre class="literal-block">
ip link set master vrf_external dev eth0
</pre>
<p>Any prefixes configured on <code>eth0</code> will be moved from the :code`main`
routing table into routing table <cite>1023</cite></p>
</div>
<div class="section" id="vrf-setup-with-ifupdown">
<h3>VRF setup with ifupdown</h3>
<p>For Debian based Distributions <code>ifupdown2</code> provides an almost drop-in
replacement for <code>ifupdown</code> including VRFs and other features.
A configuration for an interface <code>eth0</code> being part of VRF
code:<cite>vrf_external</cite> could look like this:</p>
<pre class="literal-block">
auto eth0
iface eth0
address 192.0.2.42/24
address 2001:db8:08:15::42/64
gateway 192.0.2.1
gateway 2001:db8:08:15::1
vrf vrf_external
auto vrf_external
iface vrf_external
vrf-table 1023
</pre>
</div>
<div class="section" id="openvpn-configuration">
<h3>OpenVPN configuration</h3>
<p>The OpenVPN configuration needs to contain this line:</p>
<pre class="literal-block">
bind-dev vrf_external
</pre>
</div>
<div class="section" id="further-reading">
<h3>Further reading</h3>
<p>Wikipedia has nice page one VRFs: <a class="reference external" href="https://en.wikipedia.org/wiki/Virtual_routing_and_forwarding">https://en.wikipedia.org/wiki/Virtual_routing_and_forwarding</a></p>
<p>This talk from the Network Track of FrOSCon 2018 provides an overview about
advanced layer 2 and layer 3 features of Linux</p>
<blockquote>
<ul class="simple">
<li>Slides: <a class="reference external" href="https://www.slideshare.net/BarbarossaTM/l2l3-fr-fortgeschrittene-helle-und-dunkle-magie-im-linuxnetzwerkstack">https://www.slideshare.net/BarbarossaTM/l2l3-fr-fortgeschrittene-helle-und-dunkle-magie-im-linuxnetzwerkstack</a></li>
<li>Video (german): <a class="reference external" href="https://media.ccc.de/v/froscon2018-2247-l2_l3_fur_fortgeschrittene_-_helle_und_dunkle_magie_im_linux-netzwerkstack">https://media.ccc.de/v/froscon2018-2247-l2_l3_fur_fortgeschrittene_-_helle_und_dunkle_magie_im_linux-netzwerkstack</a></li>
</ul>
</blockquote>
</div>
</div>
</div>
<div class="section" id="scripting-integration">
<h1>SCRIPTING INTEGRATION</h1>
<p>OpenVPN can execute external scripts in various phases of the lifetime of
the OpenVPN process.</p>
<div class="section" id="script-order-of-execution">
<h2>Script Order of Execution</h2>
<ol class="arabic">
<li><p class="first"><tt class="docutils literal"><span class="pre">--up</span></tt></p>
<p>Executed after TCP/UDP socket bind and TUN/TAP open.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">--tls-verify</span></tt></p>
<p>Executed when we have a still untrusted remote peer.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">--ipchange</span></tt></p>
<p>Executed after connection authentication, or remote IP address change.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">--client-connect</span></tt></p>
<p>Executed in <strong>--mode server</strong> mode immediately after client
authentication.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">--route-up</span></tt></p>
<p>Executed after connection authentication, either immediately after, or
some number of seconds after as defined by the <strong>--route-delay</strong> option.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">--route-pre-down</span></tt></p>
<p>Executed right before the routes are removed.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">--client-disconnect</span></tt></p>
<p>Executed in <tt class="docutils literal"><span class="pre">--mode</span> server</tt> mode on client instance shutdown.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">--down</span></tt></p>
<p>Executed after TCP/UDP and TUN/TAP close.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">--learn-address</span></tt></p>
<p>Executed in <tt class="docutils literal"><span class="pre">--mode</span> server</tt> mode whenever an IPv4 address/route or MAC
address is added to OpenVPN's internal routing table.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt></p>
<p>Executed in <tt class="docutils literal"><span class="pre">--mode</span> server</tt> mode on new client connections, when the
client is still untrusted.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">--client-crresponse</span></tt></p>
<blockquote>
<p>Execute in <tt class="docutils literal"><span class="pre">--mode</span> server</tt> whenever a client sends a
<code>CR_RESPONSE</code> message</p>
</blockquote>
</li>
</ol>
</div>
<div class="section" id="script-hooks">
<h2>SCRIPT HOOKS</h2>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--auth-user-pass-verify <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Require the client to provide a username/password (possibly in addition
to a client certificate) for authentication.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
auth-user-pass-verify cmd method
</pre>
<p>OpenVPN will run command <tt class="docutils literal">cmd</tt> to validate the username/password
provided by the client.</p>
<p><tt class="docutils literal">cmd</tt> consists of a path to a script (or executable program), optionally
followed by arguments. The path and arguments may be single- or
double-quoted and/or escaped using a backslash, and should be separated
by one or more spaces.</p>
<p>If <tt class="docutils literal">method</tt> is set to <code>via-env</code>, OpenVPN will call <tt class="docutils literal">cmd</tt>
with the environmental variables <code>username</code> and <code>password</code>
set to the username/password strings provided by the client. <em>Beware</em>
that this method is insecure on some platforms which make the environment
of a process publicly visible to other unprivileged processes.</p>
<p>If <tt class="docutils literal">method</tt> is set to <code>via-file</code>, OpenVPN will write the username
and password to the first two lines of a temporary file. The filename
will be passed as an argument to <tt class="docutils literal">cmd</tt>, and the file will be
automatically deleted by OpenVPN after the script returns. The location
of the temporary file is controlled by the <tt class="docutils literal"><span class="pre">--tmp-dir</span></tt> option, and
will default to the current directory if unspecified. For security,
consider setting <tt class="docutils literal"><span class="pre">--tmp-dir</span></tt> to a volatile storage medium such as
<code>/dev/shm</code> (if available) to prevent the username/password file
from touching the hard drive.</p>
<p>The script should examine the username and password, returning a success
exit code (<code>0</code>) if the client's authentication request is to be
accepted, a failure code (<code>1</code>) to reject the client, or a that
the authentication is deferred (<code>2</code>). If the authentication is
deferred, the script must fork/start a background or another non-blocking
operation to continue the authentication in the background. When finshing
the authentication, a <code>1</code> or <code>0</code> must be written to the
file specified by the <code>auth_control_file</code>.</p>
<p>If the file specified by <code>auth_failed_reason_file</code> exists and has
non-empty content, the content of this file will be used as AUTH_FAILED
message. To avoid race conditions, this file should be written before
<code>auth_control_file</code>.</p>
<p>This auth fail reason can be something simple like "User has been permanently
disabled" but there are also some special auth failed messages.</p>
<p>The <tt class="docutils literal">TEMP</tt> message indicates that the authentication
temporarily failed and that the client should continue to retry to connect.
The server can optionally give a user readable message and hint the client a
behavior how to proceed. The keywords of the <tt class="docutils literal">AUTH_FAILED,TEMP</tt> message
are comma separated keys/values and provide a hint to the client how to
proceed. Currently defined keywords are:</p>
<dl class="docutils">
<dt><tt class="docutils literal">backoff</tt> <code>s</code></dt>
<dd>instructs the client to wait at least <code>s</code> seconds before the next
connection attempt. If the client already uses a higher delay for
reconnection attempt, the delay will not be shortened.</dd>
<dt><tt class="docutils literal">advance addr</tt></dt>
<dd>Instructs the client to reconnect to the next (IP) address of the
current server.</dd>
<dt><tt class="docutils literal">advance remote</tt></dt>
<dd>Instructs the client to skip the remaining IP addresses of the current
server and instead connect to the next server specified in the
configuration file.</dd>
<dt><tt class="docutils literal">advance no</tt></dt>
<dd>Instructs the client to retry connecting to the same server again.</dd>
</dl>
<p>For example, the message <tt class="docutils literal">TEMP[backoff 42,advance no]: No free IP addresses</tt>
indicates that the VPN connection can currently not succeed and instructs
the client to retry in 42 seconds again.</p>
<p>When deferred authentication is in use, the script can also request
pending authentication by writing to the file specified by the
<code>auth_pending_file</code>. The first line must be the timeout in
seconds, the required method on the second line (e.g. crtext) and
third line must be the EXTRA as documented in the
<tt class="docutils literal"><span class="pre">client-pending-auth</span></tt> section of <cite>doc/management.txt</cite>.</p>
<p>This directive is designed to enable a plugin-style interface for
extending OpenVPN's authentication capabilities.</p>
<p>To protect against a client passing a maliciously formed username or
password string, the username string must consist only of these
characters: alphanumeric, underbar ('<code>_</code>'), dash ('<code>-</code>'),
dot ('<code>.</code>'), or at ('<code>@</code>'). The password string can consist
of any printable characters except for CR or LF. Any illegal characters
in either the username or password string will be converted to
underbar ('<code>_</code>').</p>
<p>Care must be taken by any user-defined scripts to avoid creating a
security vulnerability in the way that these strings are handled. Never
use these strings in such a way that they might be escaped or evaluated
by a shell interpreter.</p>
<p class="last">For a sample script that performs PAM authentication, see
<code>sample-scripts/auth-pam.pl</code> in the OpenVPN source distribution.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--client-crresponse</span></kbd></td>
</tr>
<tr><td> </td><td><blockquote class="first">
<p>Executed when the client sends a text based challenge response.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
client-crresponse cmd
</pre>
</blockquote>
<p>OpenVPN will write the response of the client into a temporary file.
The filename will be passed as an argument to <tt class="docutils literal">cmd</tt>, and the file will be
automatically deleted by OpenVPN after the script returns.</p>
<p>The response is passed as is from the client. The script needs to check
itself if the input is valid, e.g. if the input is valid base64 encoding.</p>
<p>The script can either directly write the result of the verification to
<code>auth_control_file or further defer it. See `</code>--auth-user-pass-verify``
for details.</p>
<p class="last">For a sample script that implement TOTP (RFC 6238) based two-factor
authentication, see <code>sample-scripts/totpauth.py</code>.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--client-connect <var>cmd</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Run command <tt class="docutils literal">cmd</tt> on client connection.</p>
<p><tt class="docutils literal">cmd</tt> consists of a path to a script (or executable program), optionally
followed by arguments. The path and arguments may be single- or
double-quoted and/or escaped using a backslash, and should be separated
by one or more spaces.</p>
<p>The command is passed the common name and IP address of the
just-authenticated client as environmental variables (see environmental
variable section below). The command is also passed the pathname of a
freshly created temporary file as the last argument (after any arguments
specified in <tt class="docutils literal">cmd</tt> ), to be used by the command to pass dynamically
generated config file directives back to OpenVPN.</p>
<p>If the script wants to generate a dynamic config file to be applied on
the server when the client connects, it should write it to the file
named by the last argument.</p>
<p>See the <tt class="docutils literal"><span class="pre">--client-config-dir</span></tt> option below for options which can be
legally used in a dynamically generated config file.</p>
<p>Note that the return value of <tt class="docutils literal">script</tt> is significant. If <tt class="docutils literal">script</tt>
returns a non-zero error status, it will cause the client to be
disconnected.</p>
<p class="last">If a <tt class="docutils literal"><span class="pre">--client-connect</span></tt> wants to defer the generating of the
configuration then the script needs to use the
<code>client_connect_deferred_file</code> and
<code>client_connect_config_file</code> environment variables, and write
status accordingly into these files. See the <a class="reference internal" href="#environmental-variables">Environmental Variables</a>
section for more details.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--client-disconnect <var>cmd</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Like <tt class="docutils literal"><span class="pre">--client-connect</span></tt> but called on client instance shutdown. Will
not be called unless the <tt class="docutils literal"><span class="pre">--client-connect</span></tt> script and plugins (if
defined) were previously called on this instance with successful (0)
status returns.</p>
<p>The exception to this rule is if the <tt class="docutils literal"><span class="pre">--client-disconnect</span></tt> command or
plugins are cascaded, and at least one client-connect function
succeeded, then ALL of the client-disconnect functions for scripts and
plugins will be called on client instance object deletion, even in cases
where some of the related client-connect functions returned an error
status.</p>
<p class="last">The <tt class="docutils literal"><span class="pre">--client-disconnect</span></tt> command is not passed any extra arguments
(only those arguments specified in cmd, if any).</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--down <var>cmd</var></span></kbd></td>
<td><p class="first">Run command <tt class="docutils literal">cmd</tt> after TUN/TAP device close (post <tt class="docutils literal"><span class="pre">--user</span></tt> UID
change and/or <tt class="docutils literal"><span class="pre">--chroot</span></tt> ). <tt class="docutils literal">cmd</tt> consists of a path to script (or
executable program), optionally followed by arguments. The path and
arguments may be single- or double-quoted and/or escaped using a
backslash, and should be separated by one or more spaces.</p>
<p>Called with the same parameters and environmental variables as the
<tt class="docutils literal"><span class="pre">--up</span></tt> option above.</p>
<p class="last">Note that if you reduce privileges by using <tt class="docutils literal"><span class="pre">--user</span></tt> and/or
<tt class="docutils literal"><span class="pre">--group</span></tt>, your <tt class="docutils literal"><span class="pre">--down</span></tt> script will also run at reduced privilege.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--down-pre</span></kbd></td>
<td>Call <tt class="docutils literal"><span class="pre">--down</span></tt> cmd/script before, rather than after, TUN/TAP close.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--ipchange <var>cmd</var></span></kbd></td>
<td><p class="first">Run command <tt class="docutils literal">cmd</tt> when our remote ip-address is initially
authenticated or changes.</p>
<p><tt class="docutils literal">cmd</tt> consists of a path to a script (or executable program), optionally
followed by arguments. The path and arguments may be single- or
double-quoted and/or escaped using a backslash, and should be separated
by one or more spaces.</p>
<p>When <tt class="docutils literal">cmd</tt> is executed two arguments are appended after any arguments
specified in <tt class="docutils literal">cmd</tt> , as follows:</p>
<pre class="literal-block">
cmd ip address port number
</pre>
<p>Don't use <tt class="docutils literal"><span class="pre">--ipchange</span></tt> in <tt class="docutils literal"><span class="pre">--mode</span> server</tt> mode. Use a
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> script instead.</p>
<p>See the <a class="reference internal" href="#environmental-variables">Environmental Variables</a> section below for additional
parameters passed as environmental variables.</p>
<p>If you are running in a dynamic IP address environment where the IP
addresses of either peer could change without notice, you can use this
script, for example, to edit the <code>/etc/hosts</code> file with the current
address of the peer. The script will be run every time the remote peer
changes its IP address.</p>
<p class="last">Similarly if <em>our</em> IP address changes due to DHCP, we should configure
our IP address change script (see man page for <tt class="docutils literal">dhcpcd</tt>(8)) to
deliver a <tt class="docutils literal">SIGHUP</tt> or <tt class="docutils literal">SIGUSR1</tt> signal to OpenVPN. OpenVPN will
then re-establish a connection with its most recently authenticated
peer on its new IP address.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--learn-address <var>cmd</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Run command <tt class="docutils literal">cmd</tt> to validate client virtual addresses or routes.</p>
<p><tt class="docutils literal">cmd</tt> consists of a path to a script (or executable program), optionally
followed by arguments. The path and arguments may be single- or
double-quoted and/or escaped using a backslash, and should be separated
by one or more spaces.</p>
<p>Three arguments will be appended to any arguments in <tt class="docutils literal">cmd</tt> as follows:</p>
<dl class="docutils">
<dt><code>$1</code> - [operation]</dt>
<dd><code>"add"</code>, <code>"update"</code>, or <code>"delete"</code> based on whether
or not the address is being added to, modified, or deleted from
OpenVPN's internal routing table.</dd>
<dt><code>$2</code> - [address]</dt>
<dd>The address being learned or unlearned. This can be an IPv4 address
such as <code>"198.162.10.14"</code>, an IPv4 subnet such as
<code>"198.162.10.0/24"</code>, or an ethernet MAC address (when
<tt class="docutils literal"><span class="pre">--dev</span> tap</tt> is being used) such as <code>"00:FF:01:02:03:04"</code>.</dd>
<dt><code>$3</code> - [common name]</dt>
<dd>The common name on the certificate associated with the client linked
to this address. Only present for <code>"add"</code> or <code>"update"</code>
operations, not <code>"delete"</code>.</dd>
</dl>
<p>On <code>"add"</code> or <code>"update"</code> methods, if the script returns
a failure code (non-zero), OpenVPN will reject the address and will not
modify its internal routing table.</p>
<p class="last">Normally, the <tt class="docutils literal">cmd</tt> script will use the information provided above to
set appropriate firewall entries on the VPN TUN/TAP interface. Since
OpenVPN provides the association between virtual IP or MAC address and
the client's authenticated common name, it allows a user-defined script
to configure firewall access policies with regard to the client's
high-level common name, rather than the low level client virtual
addresses.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--route-up <var>cmd</var></span></kbd></td>
<td><p class="first">Run command <tt class="docutils literal">cmd</tt> after routes are added, subject to <tt class="docutils literal"><span class="pre">--route-delay</span></tt>.</p>
<p><tt class="docutils literal">cmd</tt> consists of a path to a script (or executable program), optionally
followed by arguments. The path and arguments may be single- or
double-quoted and/or escaped using a backslash, and should be separated
by one or more spaces.</p>
<p class="last">See the <a class="reference internal" href="#environmental-variables">Environmental Variables</a> section below for additional
parameters passed as environmental variables.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--route-pre-down <var>cmd</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Run command <tt class="docutils literal">cmd</tt> before routes are removed upon disconnection.</p>
<p><tt class="docutils literal">cmd</tt> consists of a path to a script (or executable program), optionally
followed by arguments. The path and arguments may be single- or
double-quoted and/or escaped using a backslash, and should be separated
by one or more spaces.</p>
<p class="last">See the <a class="reference internal" href="#environmental-variables">Environmental Variables</a> section below for additional
parameters passed as environmental variables.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--setenv <var>args</var></span></kbd></td>
<td><p class="first">Set a custom environmental variable <code>name=value</code> to pass to script.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
setenv name value
setenv FORWARD_COMPATIBLE 1
setenv opt config_option
</pre>
<p>By setting <code>FORWARD_COMPATIBLE</code> to <code>1</code>, the config file
syntax checking is relaxed so that unknown directives will trigger a
warning but not a fatal error, on the assumption that a given unknown
directive might be valid in future OpenVPN versions.</p>
<p>This option should be used with caution, as there are good security
reasons for having OpenVPN fail if it detects problems in a config file.
Having said that, there are valid reasons for wanting new software
features to gracefully degrade when encountered by older software
versions.</p>
<p>It is also possible to tag a single directive so as not to trigger a
fatal error if the directive isn't recognized. To do this, prepend the
following before the directive: <tt class="docutils literal">setenv opt</tt></p>
<p>Versions prior to OpenVPN 2.3.3 will always ignore options set with the
<tt class="docutils literal">setenv opt</tt> directive.</p>
<p class="last">See also <tt class="docutils literal"><span class="pre">--ignore-unknown-option</span></tt></p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--setenv-safe <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Set a custom environmental variable <code>OPENVPN_name</code> to <code>value</code>
to pass to scripts.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
setenv-safe name value
</pre>
<p class="last">This directive is designed to be pushed by the server to clients, and
the prepending of <code>OPENVPN_</code> to the environmental variable is a
safety precaution to prevent a <code>LD_PRELOAD</code> style attack from a
malicious or compromised server.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-verify <var>cmd</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Run command <tt class="docutils literal">cmd</tt> to verify the X509 name of a pending TLS connection
that has otherwise passed all other tests of certification (except for
revocation via <tt class="docutils literal"><span class="pre">--crl-verify</span></tt> directive; the revocation test occurs
after the <tt class="docutils literal"><span class="pre">--tls-verify</span></tt> test).</p>
<p><tt class="docutils literal">cmd</tt> should return <code>0</code> to allow the TLS handshake to proceed,
or <code>1</code> to fail.</p>
<p><tt class="docutils literal">cmd</tt> consists of a path to a script (or executable program), optionally
followed by arguments. The path and arguments may be single- or
double-quoted and/or escaped using a backslash, and should be separated
by one or more spaces.</p>
<p>When <tt class="docutils literal">cmd</tt> is executed two arguments are appended after any arguments
specified in <tt class="docutils literal">cmd</tt>, as follows:</p>
<pre class="literal-block">
cmd certificate_depth subject
</pre>
<p>These arguments are, respectively, the current certificate depth and the
X509 subject distinguished name (dn) of the peer.</p>
<p>This feature is useful if the peer you want to trust has a certificate
which was signed by a certificate authority who also signed many other
certificates, where you don't necessarily want to trust all of them, but
rather be selective about which peer certificate you will accept. This
feature allows you to write a script which will test the X509 name on a
certificate and decide whether or not it should be accepted. For a
simple perl script which will test the common name field on the
certificate, see the file <tt class="docutils literal"><span class="pre">verify-cn</span></tt> in the OpenVPN distribution.</p>
<p class="last">See the <a class="reference internal" href="#environmental-variables">Environmental Variables</a> section below for additional
parameters passed as environmental variables.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tls-export-cert <var>dir</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Adds an environment variable <tt class="docutils literal">peer_cert</tt> when calling the
<tt class="docutils literal"><span class="pre">--tls-verify</span></tt> script or executing the OPENVPN_PLUGIN_TLS_VERIFY plugin
hook to verify the certificate.</p>
<p class="last">The environment variable contains the path to a PEM encoded certificate
of the current peer certificate in the directory <tt class="docutils literal">dir</tt>.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--up <var>cmd</var></span></kbd></td>
<td><p class="first">Run command <tt class="docutils literal">cmd</tt> after successful TUN/TAP device open (pre <tt class="docutils literal"><span class="pre">--user</span></tt>
UID change).</p>
<p><tt class="docutils literal">cmd</tt> consists of a path to a script (or executable program), optionally
followed by arguments. The path and arguments may be single- or
double-quoted and/or escaped using a backslash, and should be separated
by one or more spaces.</p>
<p>The up command is useful for specifying route commands which route IP
traffic destined for private subnets which exist at the other end of the
VPN connection into the tunnel.</p>
<p>For <tt class="docutils literal"><span class="pre">--dev</span> tun</tt> execute as:</p>
<pre class="literal-block">
cmd tun_dev tun_mtu 0 ifconfig_local_ip ifconfig_remote_ip [init | restart]
</pre>
<p>For <tt class="docutils literal"><span class="pre">--dev</span> tap</tt> execute as:</p>
<pre class="literal-block">
cmd tap_dev tap_mtu 0 ifconfig_local_ip ifconfig_netmask [init | restart]
</pre>
<p>See the <a class="reference internal" href="#environmental-variables">Environmental Variables</a> section below for additional
parameters passed as environmental variables. The <tt class="docutils literal">0</tt> argument
used to be <tt class="docutils literal">link_mtu</tt> which is no longer passed to scripts - to
keep the argument order, it was replaced with <tt class="docutils literal">0</tt>.</p>
<p>Note that if <tt class="docutils literal">cmd</tt> includes arguments, all OpenVPN-generated arguments
will be appended to them to build an argument list with which the
executable will be called.</p>
<p>Typically, <tt class="docutils literal">cmd</tt> will run a script to add routes to the tunnel.</p>
<p>Normally the up script is called after the TUN/TAP device is opened. In
this context, the last command line parameter passed to the script will
be <em>init.</em> If the <tt class="docutils literal"><span class="pre">--up-restart</span></tt> option is also used, the up script
will be called for restarts as well. A restart is considered to be a
partial reinitialization of OpenVPN where the TUN/TAP instance is
preserved (the <tt class="docutils literal"><span class="pre">--persist-tun</span></tt> option will enable such preservation).
A restart can be generated by a SIGUSR1 signal, a <tt class="docutils literal"><span class="pre">--ping-restart</span></tt>
timeout, or a connection reset when the TCP protocol is enabled with the
<tt class="docutils literal"><span class="pre">--proto</span></tt> option. If a restart occurs, and <tt class="docutils literal"><span class="pre">--up-restart</span></tt> has been
specified, the up script will be called with <em>restart</em> as the last
parameter.</p>
<dl class="docutils">
<dt><em>NOTE:</em></dt>
<dd>On restart, OpenVPN will not pass the full set of environment
variables to the script. Namely, everything related to routing and
gateways will not be passed, as nothing needs to be done anyway - all
the routing setup is already in place. Additionally, the up-restart
script will run with the downgraded UID/GID settings (if configured).</dd>
</dl>
<p>The following standalone example shows how the <tt class="docutils literal"><span class="pre">--up</span></tt> script can be
called in both an initialization and restart context. (<em>NOTE:</em> for
security reasons, don't run the following example unless UDP port 9999
is blocked by your firewall. Also, the example will run indefinitely, so
you should abort with control-c).</p>
<pre class="literal-block">
openvpn --dev tun --port 9999 --verb 4 --ping-restart 10 \
--up 'echo up' --down 'echo down' --persist-tun \
--up-restart
</pre>
<p>Note that OpenVPN also provides the <tt class="docutils literal"><span class="pre">--ifconfig</span></tt> option to
automatically ifconfig the TUN device, eliminating the need to define an
<tt class="docutils literal"><span class="pre">--up</span></tt> script, unless you also want to configure routes in the
<tt class="docutils literal"><span class="pre">--up</span></tt> script.</p>
<p>If <tt class="docutils literal"><span class="pre">--ifconfig</span></tt> is also specified, OpenVPN will pass the ifconfig
local and remote endpoints on the command line to the <tt class="docutils literal"><span class="pre">--up</span></tt> script so
that they can be used to configure routes such as:</p>
<pre class="last literal-block">
route add -net 10.0.0.0 netmask 255.255.255.0 gw $5
</pre>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--up-delay</span></kbd></td>
<td><p class="first">Delay TUN/TAP open and possible <tt class="docutils literal"><span class="pre">--up</span></tt> script execution until after
TCP/UDP connection establishment with peer.</p>
<p>In <tt class="docutils literal"><span class="pre">--proto</span> udp</tt> mode, this option normally requires the use of
<tt class="docutils literal"><span class="pre">--ping</span></tt> to allow connection initiation to be sensed in the absence of
tunnel data, since UDP is a "connectionless" protocol.</p>
<p class="last">On Windows, this option will delay the TAP-Win32 media state
transitioning to "connected" until connection establishment, i.e. the
receipt of the first authenticated packet from the peer.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--up-restart</span></kbd></td>
<td>Enable the <tt class="docutils literal"><span class="pre">--up</span></tt> and <tt class="docutils literal"><span class="pre">--down</span></tt> scripts to be called for restarts as
well as initial program start. This option is described more fully above
in the <tt class="docutils literal"><span class="pre">--up</span></tt> option documentation.</td></tr>
</tbody>
</table>
</div>
<div class="section" id="string-types-and-remapping">
<h2>String Types and Remapping</h2>
<p>In certain cases, OpenVPN will perform remapping of characters in
strings. Essentially, any characters outside the set of permitted
characters for each string type will be converted to underbar ('_').</p>
<dl class="docutils">
<dt><em>Q: Why is string remapping necessary?</em></dt>
<dd>It's an important security feature to prevent the malicious
coding of strings from untrusted sources to be passed as parameters to
scripts, saved in the environment, used as a common name, translated to
a filename, etc.</dd>
<dt><em>Q: Can string remapping be disabled?</em></dt>
<dd>Yes, by using the <tt class="docutils literal"><span class="pre">--no-name-remapping</span></tt> option, however this
should be considered an advanced option.</dd>
</dl>
<p>Here is a brief rundown of OpenVPN's current string types and the
permitted character class for each string:</p>
<dl class="docutils">
<dt><em>X509 Names</em></dt>
<dd>Alphanumeric, underbar ('_'), dash ('-'), dot ('.'), at
('@'), colon (':'), slash ('/'), and equal ('='). Alphanumeric is
defined as a character which will cause the C library isalnum() function
to return true.</dd>
<dt><em>Common Names</em></dt>
<dd>Alphanumeric, underbar ('_'), dash ('-'), dot ('.'), and at ('@').</dd>
<dt><em>--auth-user-pass username</em></dt>
<dd>Same as Common Name, with one exception:
starting with OpenVPN 2.0.1, the username is passed to the
<code>OPENVPN_PLUGIN_AUTH_USER_PASS_VERIFY</code> plugin in its raw form,
without string remapping.</dd>
<dt><em>--auth-user-pass password</em></dt>
<dd>Any "printable" character except CR or LF. Printable is defined to be
a character which will cause the C library isprint() function to
return true.</dd>
<dt><em>--client-config-dir filename as derived from common name or`username</em></dt>
<dd>Alphanumeric, underbar ('_'), dash ('-'), and dot ('.') except for "."
or ".." as standalone strings. As of v2.0.1-rc6, the at ('@') character
has been added as well for compatibility with the common name character
class.</dd>
<dt><em>Environmental variable names</em></dt>
<dd>Alphanumeric or underbar ('_').</dd>
<dt><em>Environmental variable values</em></dt>
<dd>Any printable character.</dd>
</dl>
<p>For all cases, characters in a string which are not members of the legal
character class for that string type will be remapped to underbar
('_').</p>
</div>
<div class="section" id="environmental-variables">
<h2>Environmental Variables</h2>
<p>Once set, a variable is persisted indefinitely until it is reset by a
new value or a restart,</p>
<p>As of OpenVPN 2.0-beta12, in server mode, environmental variables set by
OpenVPN are scoped according to the client objects they are associated
with, so there should not be any issues with scripts having access to
stale, previously set variables which refer to different client
instances.</p>
<dl class="docutils">
<dt><code>bytes_received</code></dt>
<dd>Total number of bytes received from client during VPN session. Set prior
to execution of the <tt class="docutils literal"><span class="pre">--client-disconnect</span></tt> script.</dd>
<dt><code>bytes_sent</code></dt>
<dd>Total number of bytes sent to client during VPN session. Set prior to
execution of the <tt class="docutils literal"><span class="pre">--client-disconnect</span></tt> script.</dd>
<dt><code>client_connect_config_file</code></dt>
<dd>The path to the configuration file that should be written to by the
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> script (optional, if per-session configuration
is desired). This is the same file name as passed via command line
argument on the call to the <tt class="docutils literal"><span class="pre">--client-connect</span></tt> script.</dd>
<dt><code>client_connect_deferred_file</code></dt>
<dd><p class="first">This file can be optionally written to in order to to communicate a
status code of the <tt class="docutils literal"><span class="pre">--client-connect</span></tt> script or plgin. Only the
first character in the file is relevant. It must be either <code>1</code>
to indicate normal script execution, <code>0</code> indicates an error (in
the same way that a non zero exit status does) or <code>2</code> to indicate
that the script deferred returning the config file.</p>
<p>For deferred (background) handling, the script or plugin MUST write
<code>2</code> to the file to indicate the deferral and then return with
exit code <code>0</code> to signal <tt class="docutils literal">deferred handler started OK</tt>.</p>
<p>A background process or similar must then take care of writing the
configuration to the file indicated by the
<code>client_connect_config_file</code> environment variable and when
finished, write the a <code>1</code> to this file (or <code>0</code> in case of
an error).</p>
<p class="last">The absence of any character in the file when the script finishes
executing is interpreted the same as <code>1</code>. This allows scripts
that are not written to support the defer mechanism to be used
unmodified.</p>
</dd>
<dt><code>common_name</code></dt>
<dd>The X509 common name of an authenticated client. Set prior to execution
of <tt class="docutils literal"><span class="pre">--client-connect</span></tt>, <tt class="docutils literal"><span class="pre">--client-disconnect</span></tt> and
<tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> scripts.</dd>
<dt><code>config</code></dt>
<dd>Name of first <tt class="docutils literal"><span class="pre">--config</span></tt> file. Set on program initiation and reset on
SIGHUP.</dd>
<dt><code>daemon</code></dt>
<dd>Set to "1" if the <tt class="docutils literal"><span class="pre">--daemon</span></tt> directive is specified, or "0" otherwise.
Set on program initiation and reset on SIGHUP.</dd>
<dt><code>daemon_log_redirect</code></dt>
<dd>Set to "1" if the <tt class="docutils literal"><span class="pre">--log</span></tt> or <tt class="docutils literal"><span class="pre">--log-append</span></tt> directives are
specified, or "0" otherwise. Set on program initiation and reset on
SIGHUP.</dd>
<dt><code>dev</code></dt>
<dd>The actual name of the TUN/TAP device, including a unit number if it
exists. Set prior to <tt class="docutils literal"><span class="pre">--up</span></tt> or <tt class="docutils literal"><span class="pre">--down</span></tt> script execution.</dd>
<dt><code>dev_idx</code></dt>
<dd>On Windows, the device index of the TUN/TAP adapter (to be used in
netsh.exe calls which sometimes just do not work right with interface
names). Set prior to <tt class="docutils literal"><span class="pre">--up</span></tt> or <tt class="docutils literal"><span class="pre">--down</span></tt> script execution.</dd>
<dt><code>dns_*</code></dt>
<dd><p class="first">The <tt class="docutils literal"><span class="pre">--dns</span></tt> configuration options will be made available to script
execution through this set of environment variables. Variables appear
only if the corresponding option has a value assigned. For the semantics
of each individual variable, please refer to the documentation for <tt class="docutils literal"><span class="pre">--dns</span></tt>.</p>
<pre class="last literal-block">
dns_search_domain_{n}
dns_server_{n}_address_{m}
dns_server_{n}_port_{m}
dns_server_{n}_resolve_domain_{m}
dns_server_{n}_dnssec
dns_server_{n}_transport
dns_server_{n}_sni
</pre>
</dd>
<dt><code>foreign_option_{n}</code></dt>
<dd>An option pushed via <tt class="docutils literal"><span class="pre">--push</span></tt> to a client which does not natively
support it, such as <tt class="docutils literal"><span class="pre">--dhcp-option</span></tt> on a non-Windows system, will be
recorded to this environmental variable sequence prior to <tt class="docutils literal"><span class="pre">--up</span></tt>
script execution.</dd>
<dt><code>ifconfig_broadcast</code></dt>
<dd>The broadcast address for the virtual ethernet segment which is derived
from the <tt class="docutils literal"><span class="pre">--ifconfig</span></tt> option when <tt class="docutils literal"><span class="pre">--dev</span> tap</tt> is used. Set prior to
OpenVPN calling the <code>ifconfig</code> or <code>netsh</code> (windows version
of ifconfig) commands which normally occurs prior to <tt class="docutils literal"><span class="pre">--up</span></tt> script
execution.</dd>
<dt><code>ifconfig_ipv6_local</code></dt>
<dd>The local VPN endpoint IPv6 address specified in the
<tt class="docutils literal"><span class="pre">--ifconfig-ipv6</span></tt> option (first parameter). Set prior to OpenVPN
calling the <code>ifconfig</code> or code:<cite>netsh</cite> (windows version of
ifconfig) commands which normally occurs prior to <tt class="docutils literal"><span class="pre">--up</span></tt> script
execution.</dd>
<dt><code>ifconfig_ipv6_netbits</code></dt>
<dd>The prefix length of the IPv6 network on the VPN interface. Derived
from the /nnn parameter of the IPv6 address in the <tt class="docutils literal"><span class="pre">--ifconfig-ipv6</span></tt>
option (first parameter). Set prior to OpenVPN calling the
<code>ifconfig</code> or <code>netsh</code> (windows version of ifconfig)
commands which normally occurs prior to <tt class="docutils literal"><span class="pre">--up</span></tt> script execution.</dd>
<dt><code>ifconfig_ipv6_remote</code></dt>
<dd>The remote VPN endpoint IPv6 address specified in the
<tt class="docutils literal"><span class="pre">--ifconfig-ipv6</span></tt> option (second parameter). Set prior to OpenVPN
calling the <code>ifconfig</code> or <code>netsh</code> (windows version of
ifconfig) commands which normally occurs prior to <tt class="docutils literal"><span class="pre">--up</span></tt> script
execution.</dd>
<dt><code>ifconfig_local</code></dt>
<dd>The local VPN endpoint IP address specified in the <tt class="docutils literal"><span class="pre">--ifconfig</span></tt>
option (first parameter). Set prior to OpenVPN calling the
<code>ifconfig</code> or <code>netsh</code> (windows version of ifconfig)
commands which normally occurs prior to <tt class="docutils literal"><span class="pre">--up</span></tt> script execution.</dd>
<dt><code>ifconfig_remote</code></dt>
<dd>The remote VPN endpoint IP address specified in the <tt class="docutils literal"><span class="pre">--ifconfig</span></tt>
option (second parameter) when <tt class="docutils literal"><span class="pre">--dev</span> tun</tt> is used. Set prior to
OpenVPN calling the <code>ifconfig</code> or <code>netsh</code> (windows version
of ifconfig) commands which normally occurs prior to <tt class="docutils literal"><span class="pre">--up</span></tt> script
execution.</dd>
<dt><code>ifconfig_netmask</code></dt>
<dd>The subnet mask of the virtual ethernet segment that is specified as
the second parameter to <tt class="docutils literal"><span class="pre">--ifconfig</span></tt> when <tt class="docutils literal"><span class="pre">--dev</span> tap</tt> is being
used. Set prior to OpenVPN calling the <code>ifconfig</code> or
<code>netsh</code> (windows version of ifconfig) commands which normally
occurs prior to <tt class="docutils literal"><span class="pre">--up</span></tt> script execution.</dd>
<dt><code>ifconfig_pool_local_ip</code></dt>
<dd>The local virtual IP address for the TUN/TAP tunnel taken from an
<tt class="docutils literal"><span class="pre">--ifconfig-push</span></tt> directive if specified, or otherwise from the
ifconfig pool (controlled by the <tt class="docutils literal"><span class="pre">--ifconfig-pool</span></tt> config file
directive). Only set for <tt class="docutils literal"><span class="pre">--dev</span> tun</tt> tunnels. This option is set on
the server prior to execution of the <tt class="docutils literal"><span class="pre">--client-connect</span></tt> and
<tt class="docutils literal"><span class="pre">--client-disconnect</span></tt> scripts.</dd>
<dt><code>ifconfig_pool_netmask</code></dt>
<dd>The virtual IP netmask for the TUN/TAP tunnel taken from an
<tt class="docutils literal"><span class="pre">--ifconfig-push</span></tt> directive if specified, or otherwise from the
ifconfig pool (controlled by the <tt class="docutils literal"><span class="pre">--ifconfig-pool</span></tt> config file
directive). Only set for <tt class="docutils literal"><span class="pre">--dev</span> tap</tt> tunnels. This option is set on
the server prior to execution of the <tt class="docutils literal"><span class="pre">--client-connect</span></tt> and
<tt class="docutils literal"><span class="pre">--client-disconnect</span></tt> scripts.</dd>
<dt><code>ifconfig_pool_remote_ip</code></dt>
<dd>The remote virtual IP address for the TUN/TAP tunnel taken from an
<tt class="docutils literal"><span class="pre">--ifconfig-push</span></tt> directive if specified, or otherwise from the
ifconfig pool (controlled by the <tt class="docutils literal"><span class="pre">--ifconfig-pool</span></tt> config file
directive). This option is set on the server prior to execution of the
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> and <tt class="docutils literal"><span class="pre">--client-disconnect</span></tt> scripts.</dd>
<dt><code>link_mtu</code></dt>
<dd>No longer passed to scripts since OpenVPN 2.6.0. Used to be the
maximum packet size (not including the IP header) of tunnel data in
UDP tunnel transport mode.</dd>
<dt><code>local</code></dt>
<dd>The <tt class="docutils literal"><span class="pre">--local</span></tt> parameter. Set on program initiation and reset on
SIGHUP.</dd>
<dt><code>local_port</code></dt>
<dd>The local port number or name, specified by <tt class="docutils literal"><span class="pre">--port</span></tt> or <tt class="docutils literal"><span class="pre">--lport</span></tt>.
Set on program initiation and reset on SIGHUP.</dd>
<dt><code>password</code></dt>
<dd>The password provided by a connecting client. Set prior to
<tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> script execution only when the <tt class="docutils literal"><span class="pre">via-env</span></tt>
modifier is specified, and deleted from the environment after the script
returns.</dd>
<dt><code>peer_cert</code></dt>
<dd>If the option <tt class="docutils literal"><span class="pre">--tls-export-cert</span></tt> is enabled, this option contains
the path to the current peer certificate to be verified in PEM format.
See also the argument certificate_depth to the <tt class="docutils literal"><span class="pre">--tls-verify</span></tt> command.</dd>
<dt><code>proto</code></dt>
<dd>The <tt class="docutils literal"><span class="pre">--proto</span></tt> parameter. Set on program initiation and reset on
SIGHUP.</dd>
<dt><code>remote_{n}</code></dt>
<dd>The <tt class="docutils literal"><span class="pre">--remote</span></tt> parameter. Set on program initiation and reset on
SIGHUP.</dd>
<dt><code>remote_port_{n}</code></dt>
<dd>The remote port number, specified by <tt class="docutils literal"><span class="pre">--port</span></tt> or <tt class="docutils literal"><span class="pre">--rport</span></tt>. Set on
program initiation and reset on SIGHUP.</dd>
<dt><code>route_net_gateway</code></dt>
<dd>The pre-existing default IP gateway in the system routing table. Set
prior to <tt class="docutils literal"><span class="pre">--up</span></tt> script execution.</dd>
<dt><code>route_vpn_gateway</code></dt>
<dd>The default gateway used by <tt class="docutils literal"><span class="pre">--route</span></tt> options, as specified in either
the <tt class="docutils literal"><span class="pre">--route-gateway</span></tt> option or the second parameter to
<tt class="docutils literal"><span class="pre">--ifconfig</span></tt> when <tt class="docutils literal"><span class="pre">--dev</span> tun</tt> is specified. Set prior to <tt class="docutils literal"><span class="pre">--up</span></tt>
script execution.</dd>
<dt><code>route_{parm}_{n}</code></dt>
<dd><p class="first">A set of variables which define each route to be added, and are set
prior to <tt class="docutils literal"><span class="pre">--up</span></tt> script execution.</p>
<p><tt class="docutils literal">parm</tt> will be one of <code>network</code>, <code>netmask"</code>,
<code>gateway</code>, or <code>metric</code>.</p>
<p><tt class="docutils literal">n</tt> is the OpenVPN route number, starting from 1.</p>
<p class="last">If the network or gateway are resolvable DNS names, their IP address
translations will be recorded rather than their names as denoted on the
command line or configuration file.</p>
</dd>
<dt><code>route_ipv6_{parm}_{n}</code></dt>
<dd><p class="first">A set of variables which define each IPv6 route to be added, and are
set prior to <strong>--up</strong> script execution.</p>
<p><tt class="docutils literal">parm</tt> will be one of <code>network</code>, <code>gateway</code> or
<code>metric</code>. <tt class="docutils literal">route_ipv6_network_{n}</tt> contains <code>netmask</code>
as <code>/nnn</code>, unlike IPv4 where it is passed in a separate environment
variable.</p>
<p><tt class="docutils literal">n</tt> is the OpenVPN route number, starting from 1.</p>
<p class="last">If the network or gateway are resolvable DNS names, their IP address
translations will be recorded rather than their names as denoted on the
command line or configuration file.</p>
</dd>
<dt><code>script_context</code></dt>
<dd>Set to "init" or "restart" prior to up/down script execution. For more
information, see documentation for <tt class="docutils literal"><span class="pre">--up</span></tt>.</dd>
<dt><code>script_type</code></dt>
<dd>Prior to execution of any script, this variable is set to the type of
script being run. It can be one of the following: <code>up</code>,
<code>down</code>, <code>ipchange</code>, <code>route-up</code>, <code>tls-verify</code>,
<code>auth-user-pass-verify</code>, <code>client-connect</code>,
<code>client-disconnect</code> or <code>learn-address</code>. Set prior to
execution of any script.</dd>
<dt><code>signal</code></dt>
<dd>The reason for exit or restart. Can be one of <code>sigusr1</code>,
<code>sighup</code>, <code>sigterm</code>, <code>sigint</code>, <code>inactive</code>
(controlled by <tt class="docutils literal"><span class="pre">--inactive</span></tt> option), <code>ping-exit</code> (controlled
by <tt class="docutils literal"><span class="pre">--ping-exit</span></tt> option), <code>ping-restart</code> (controlled by
<tt class="docutils literal"><span class="pre">--ping-restart</span></tt> option), <code>connection-reset</code> (triggered on TCP
connection reset), <code>error</code> or <code>unknown</code> (unknown signal).
This variable is set just prior to down script execution.</dd>
<dt><code>time_ascii</code></dt>
<dd>Client connection timestamp, formatted as a human-readable time string.
Set prior to execution of the <tt class="docutils literal"><span class="pre">--client-connect</span></tt> script.</dd>
<dt><code>time_duration</code></dt>
<dd>The duration (in seconds) of the client session which is now
disconnecting. Set prior to execution of the <tt class="docutils literal"><span class="pre">--client-disconnect</span></tt>
script.</dd>
<dt><code>time_unix</code></dt>
<dd>Client connection timestamp, formatted as a unix integer date/time
value. Set prior to execution of the <tt class="docutils literal"><span class="pre">--client-connect</span></tt> script.</dd>
<dt><code>tls_digest_{n}</code> / <code>tls_digest_sha256_{n}</code></dt>
<dd>Contains the certificate SHA1 / SHA256 fingerprint, where <tt class="docutils literal">n</tt> is the
verification level. Only set for TLS connections. Set prior to execution
of <tt class="docutils literal"><span class="pre">--tls-verify</span></tt> script.</dd>
<dt><code>tls_id_{n}</code></dt>
<dd>A series of certificate fields from the remote peer, where <tt class="docutils literal">n</tt> is the
verification level. Only set for TLS connections. Set prior to execution
of <tt class="docutils literal"><span class="pre">--tls-verify</span></tt> script.</dd>
<dt><code>tls_serial_{n}</code></dt>
<dd>The serial number of the certificate from the remote peer, where <tt class="docutils literal">n</tt>
is the verification level. Only set for TLS connections. Set prior to
execution of <tt class="docutils literal"><span class="pre">--tls-verify</span></tt> script. This is in the form of a decimal
string like "933971680", which is suitable for doing serial-based OCSP
queries (with OpenSSL, do not prepend "0x" to the string) If something
goes wrong while reading the value from the certificate it will be an
empty string, so your code should check that. See the
<code>contrib/OCSP_check/OCSP_check.sh</code> script for an example.</dd>
<dt><code>tls_serial_hex_{n}</code></dt>
<dd>Like <code>tls_serial_{n}</code>, but in hex form (e.g.
<code>12:34:56:78:9A</code>).</dd>
<dt><code>tun_mtu</code></dt>
<dd>The MTU of the TUN/TAP device. Set prior to <tt class="docutils literal"><span class="pre">--up</span></tt> or <tt class="docutils literal"><span class="pre">--down</span></tt>
script execution.</dd>
<dt><code>trusted_ip</code> / <code>trusted_ip6</code>)</dt>
<dd>Actual IP address of connecting client or peer which has been
authenticated. Set prior to execution of <tt class="docutils literal"><span class="pre">--ipchange</span></tt>,
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> and <tt class="docutils literal"><span class="pre">--client-disconnect</span></tt> scripts. If using ipv6
endpoints (udp6, tcp6), <code>trusted_ip6</code> will be set instead.</dd>
<dt><code>trusted_port</code></dt>
<dd>Actual port number of connecting client or peer which has been
authenticated. Set prior to execution of <tt class="docutils literal"><span class="pre">--ipchange</span></tt>,
<tt class="docutils literal"><span class="pre">--client-connect</span></tt> and <tt class="docutils literal"><span class="pre">--client-disconnect</span></tt> scripts.</dd>
<dt><code>untrusted_ip</code> / <code>untrusted_ip6</code></dt>
<dd>Actual IP address of connecting client or peer which has not been
authenticated yet. Sometimes used to <em>nmap</em> the connecting host in a
<tt class="docutils literal"><span class="pre">--tls-verify</span></tt> script to ensure it is firewalled properly. Set prior
to execution of <tt class="docutils literal"><span class="pre">--tls-verify</span></tt> and <tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt>
scripts. If using ipv6 endpoints (udp6, tcp6), <code>untrusted_ip6</code>
will be set instead.</dd>
<dt><code>untrusted_port</code></dt>
<dd>Actual port number of connecting client or peer which has not been
authenticated yet. Set prior to execution of <tt class="docutils literal"><span class="pre">--tls-verify</span></tt> and
<tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> scripts.</dd>
<dt><code>username</code></dt>
<dd>The username provided by a connecting client. Set prior to
<tt class="docutils literal"><span class="pre">--auth-user-pass-verify</span></tt> script execution only when the
<code>via-env</code> modifier is specified.</dd>
<dt><code>X509_{n}_{subject_field}</code></dt>
<dd><p class="first">An X509 subject field from the remote peer certificate, where <tt class="docutils literal">n</tt> is
the verification level. Only set for TLS connections. Set prior to
execution of <tt class="docutils literal"><span class="pre">--tls-verify</span></tt> script. This variable is similar to
<code>tls_id_{n}</code> except the component X509 subject fields are broken
out, and no string remapping occurs on these field values (except for
remapping of control characters to "<code>_</code>"). For example, the
following variables would be set on the OpenVPN server using the sample
client certificate in sample-keys (client.crt). Note that the
verification level is 0 for the client certificate and 1 for the CA
certificate.</p>
<p>You can use the <tt class="docutils literal"><span class="pre">--x509-track</span></tt> option to export more or less information
from the certificates.</p>
<pre class="last literal-block">
X509_0_emailAddress=me@myhost.mydomain
X509_0_CN=Test-Client
X509_0_O=OpenVPN-TEST
X509_0_ST=NA
X509_0_C=KG
X509_1_emailAddress=me@myhost.mydomain
X509_1_O=OpenVPN-TEST
X509_1_L=BISHKEK
X509_1_ST=NA
X509_1_C=KG
</pre>
</dd>
</dl>
</div>
<div class="section" id="management-interface-options">
<h2>Management Interface Options</h2>
<p>OpenVPN provides a feature rich socket based management interface for both
server and client mode operations.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Enable a management server on a <tt class="docutils literal"><span class="pre">socket-name</span></tt> Unix socket on those
platforms supporting it, or on a designated TCP port.</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
management socket-name unix #
management socket-name unix pw-file # (recommended)
management IP port # (INSECURE)
management IP port pw-file #
</pre>
<p><tt class="docutils literal"><span class="pre">pw-file</span></tt>, if specified, is a password file where the password must
be on first line. Instead of a filename it can use the keyword stdin
which will prompt the user for a password to use when OpenVPN is
starting.</p>
<p>For unix sockets, the default behaviour is to create a unix domain
socket that may be connected to by any process. Use the
<tt class="docutils literal"><span class="pre">--management-client-user</span></tt> and <tt class="docutils literal"><span class="pre">--management-client-group</span></tt>
directives to restrict access.</p>
<p>The management interface provides a special mode where the TCP
management link can operate over the tunnel itself. To enable this mode,
set IP to <tt class="docutils literal">tunnel</tt>. Tunnel mode will cause the management interface to
listen for a TCP connection on the local VPN address of the TUN/TAP
interface.</p>
<p><strong>*BEWARE*</strong> of enabling the management interface over TCP. In these cases
you should <em>ALWAYS</em> make use of <tt class="docutils literal"><span class="pre">pw-file</span></tt> to password protect the
management interface. Any user who can connect to this TCP <tt class="docutils literal">IP:port</tt>
will be able to manage and control (and interfere with) the OpenVPN
process. It is also strongly recommended to set IP to 127.0.0.1
(localhost) to restrict accessibility of the management server to local
clients.</p>
<p>While the management port is designed for programmatic control of
OpenVPN by other applications, it is possible to telnet to the port,
using a telnet client in "raw" mode. Once connected, type <code>help</code>
for a list of commands.</p>
<p class="last">For detailed documentation on the management interface, see the
<em>management-notes.txt</em> file in the management folder of the OpenVPN
source distribution.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-client</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Management interface will connect as a TCP/unix domain client to
<tt class="docutils literal">IP:port</tt> specified by <tt class="docutils literal"><span class="pre">--management</span></tt> rather than listen as a TCP
server or on a unix domain socket.</p>
<p class="last">If the client connection fails to connect or is disconnected, a SIGTERM
signal will be generated causing OpenVPN to quit.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-client-auth</span></kbd></td>
</tr>
<tr><td> </td><td>Gives management interface client the responsibility to authenticate
clients after their client certificate has been verified. See
<code>management-notes.txt</code> in OpenVPN distribution for detailed notes.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-client-group <var>g</var></span></kbd></td>
</tr>
<tr><td> </td><td>When the management interface is listening on a unix domain socket, only
allow connections from group <tt class="docutils literal">g</tt>.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-client-user <var>u</var></span></kbd></td>
</tr>
<tr><td> </td><td>When the management interface is listening on a unix domain socket, only
allow connections from user <tt class="docutils literal">u</tt>.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-external-cert <var>certificate-hint</var></span></kbd></td>
</tr>
<tr><td> </td><td>Allows usage for external certificate instead of <tt class="docutils literal"><span class="pre">--cert</span></tt> option
(client-only). <tt class="docutils literal"><span class="pre">certificate-hint</span></tt> is an arbitrary string which is
passed to a management interface client as an argument of
<em>NEED-CERTIFICATE</em> notification. Requires <tt class="docutils literal"><span class="pre">--management-external-key</span></tt>.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-external-key <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Allows usage for external private key file instead of <tt class="docutils literal"><span class="pre">--key</span></tt> option
(client-only).</p>
<p>Valid syntaxes:</p>
<pre class="literal-block">
management-external-key
management-external-key nopadding
management-external-key pkcs1
management-external-key pss
</pre>
<p>or any combination like:</p>
<pre class="literal-block">
management-external-key nopadding pkcs1
management-external-key pkcs1 pss
</pre>
<p class="last">The optional parameters <code>nopadding</code> <code>pkcs1</code> and <code>pss</code> signal
support for different padding algorithms. See
<code>doc/mangement-notes.txt</code> for a complete description of this
feature.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-forget-disconnect</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Make OpenVPN forget passwords when management session disconnects.</p>
<p class="last">This directive does not affect the <tt class="docutils literal"><span class="pre">--http-proxy</span></tt> username/password.
It is always cached.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-hold</span></kbd></td>
</tr>
<tr><td> </td><td>Start OpenVPN in a hibernating state, until a client of the management
interface explicitly starts it with the <code>hold release</code> command.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-log-cache <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td>Cache the most recent <tt class="docutils literal">n</tt> lines of log file history for usage by the
management channel.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-query-passwords</span></kbd></td>
</tr>
<tr><td> </td><td>Query management channel for private key password and
<tt class="docutils literal"><span class="pre">--auth-user-pass</span></tt> username/password. Only query the management
channel for inputs which ordinarily would have been queried from the
console.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-query-proxy</span></kbd></td>
</tr>
<tr><td> </td><td>Query management channel for proxy server information for a specific
<tt class="docutils literal"><span class="pre">--remote</span></tt> (client-only).</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-query-remote</span></kbd></td>
</tr>
<tr><td> </td><td>Allow management interface to override <tt class="docutils literal"><span class="pre">--remote</span></tt> directives
(client-only).</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-signal</span></kbd></td>
</tr>
<tr><td> </td><td>Send SIGUSR1 signal to OpenVPN if management session disconnects. This
is useful when you wish to disconnect an OpenVPN session on user logoff.
For <tt class="docutils literal"><span class="pre">--management-client</span></tt> this option is not needed since a disconnect
will always generate a <code>SIGTERM</code>.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-up-down</span></kbd></td>
</tr>
<tr><td> </td><td>Report tunnel up/down events to management interface.</td></tr>
</tbody>
</table>
</div>
<div class="section" id="plug-in-interface-options">
<h2>Plug-in Interface Options</h2>
<p>OpenVPN can be extended by loading external plug-in modules at runtime. These
plug-ins must be prebuilt and adhere to the OpenVPN Plug-In API.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group">
<kbd><span class="option">--plugin <var>args</var></span></kbd></td>
<td><p class="first">Loads an OpenVPN plug-in module.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
plugin module-name
plugin module-name "arguments"
</pre>
<p>The <tt class="docutils literal"><span class="pre">module-name</span></tt> needs to be the first
argument, indicating the plug-in to load. The second argument is an
optional init string which will be passed directly to the plug-in.
If the init consists of multiple arguments it must be enclosed in
double-quotes ("). Multiple plugin modules may be loaded into one
OpenVPN process.</p>
<p>The <tt class="docutils literal"><span class="pre">module-name</span></tt> argument can be just a filename or a filename
with a relative or absolute path. The format of the filename and path
defines if the plug-in will be loaded from a default plug-in directory
or outside this directory.</p>
<pre class="literal-block">
--plugin path Effective directory used
===================== =============================
myplug.so DEFAULT_DIR/myplug.so
subdir/myplug.so DEFAULT_DIR/subdir/myplug.so
./subdir/myplug.so CWD/subdir/myplug.so
/usr/lib/my/plug.so /usr/lib/my/plug.so
</pre>
<p><tt class="docutils literal">DEFAULT_DIR</tt> is replaced by the default plug-in directory, which is
configured at the build time of OpenVPN. <tt class="docutils literal">CWD</tt> is the current directory
where OpenVPN was started or the directory OpenVPN have switched into
via the <tt class="docutils literal"><span class="pre">--cd</span></tt> option before the <tt class="docutils literal"><span class="pre">--plugin</span></tt> option.</p>
<p>For more information and examples on how to build OpenVPN plug-in
modules, see the README file in the <tt class="docutils literal">plugin</tt> folder of the OpenVPN
source distribution.</p>
<p>If you are using an RPM install of OpenVPN, see
<code>/usr/share/openvpn/plugin</code>. The documentation is in <tt class="docutils literal">doc</tt> and
the actual plugin modules are in <tt class="docutils literal">lib</tt>.</p>
<p>Multiple plugin modules can be cascaded, and modules can be used in
tandem with scripts. The modules will be called by OpenVPN in the order
that they are declared in the config file. If both a plugin and script
are configured for the same callback, the script will be called last. If
the return code of the module/script controls an authentication function
(such as tls-verify, auth-user-pass-verify, or client-connect), then
every module and script must return success (<code>0</code>) in order for the
connection to be authenticated.</p>
<dl class="last docutils">
<dt><strong>WARNING</strong>:</dt>
<dd>Plug-ins may do deferred execution, meaning the plug-in will
return the control back to the main OpenVPN process and provide
the plug-in result later on via a different thread or process.
OpenVPN does <strong>NOT</strong> support multiple authentication plug-ins
<strong>where more than one plugin</strong> tries to do deferred authentication.
If this behaviour is detected, OpenVPN will shut down upon first
authentication.</dd>
</dl>
</td></tr>
</tbody>
</table>
</div>
<div class="section" id="windows-specific-options">
<h2>Windows-Specific Options</h2>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--allow-nonadmin <var>TAP-adapter</var></span></kbd></td>
</tr>
<tr><td> </td><td>(Standalone) Set <tt class="docutils literal"><span class="pre">TAP-adapter</span></tt> to allow access from non-administrative
accounts. If <tt class="docutils literal"><span class="pre">TAP-adapter</span></tt> is omitted, all TAP adapters on the system
will be configured to allow non-admin access. The non-admin access
setting will only persist for the length of time that the TAP-Win32
device object and driver remain loaded, and will need to be re-enabled
after a reboot, or if the driver is unloaded and reloaded. This
directive can only be used by an administrator.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--block-outside-dns</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Block DNS servers on other network adapters to prevent DNS leaks. This
option prevents any application from accessing TCP or UDP port 53 except
one inside the tunnel. It uses Windows Filtering Platform (WFP) and
works on Windows Vista or later.</p>
<p class="last">This option is considered unknown on non-Windows platforms and
unsupported on Windows XP, resulting in fatal error. You may want to use
<tt class="docutils literal"><span class="pre">--setenv</span> opt</tt> or <tt class="docutils literal"><span class="pre">--ignore-unknown-option</span></tt> (not suitable for
Windows XP) to ignore said error. Note that pushing unknown options from
server does not trigger fatal errors.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--cryptoapicert <var>select-string</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first"><em>(Windows/OpenSSL Only)</em> Load the certificate and private key from the
Windows Certificate System Store.</p>
<p>Use this option instead of <tt class="docutils literal"><span class="pre">--cert</span></tt> and <tt class="docutils literal"><span class="pre">--key</span></tt>.</p>
<p>This makes it possible to use any smart card, supported by Windows, but
also any kind of certificate, residing in the Cert Store, where you have
access to the private key. This option has been tested with a couple of
different smart cards (GemSAFE, Cryptoflex, and Swedish Post Office eID)
on the client side, and also an imported PKCS12 software certificate on
the server side.</p>
<p>To select a certificate, based on a substring search in the
certificate's subject:</p>
<pre class="literal-block">
cryptoapicert "SUBJ:Peter Runestig"
</pre>
<p>To select a certificate, based on certificate's thumbprint (SHA1 hash):</p>
<pre class="literal-block">
cryptoapicert "THUMB:f6 49 24 41 01 b4 ..."
</pre>
<p>The thumbprint hex string can easily be copy-and-pasted from the Windows
Certificate Store GUI. The embedded spaces in the hex string are optional.</p>
<p>To select a certificate based on a substring in certificate's
issuer name:</p>
<pre class="literal-block">
cryptoapicert "ISSUER:Sample CA"
</pre>
<p class="last">The first non-expired certificate found in the user's store or the
machine store that matches the select-string is used.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--dhcp-release</span></kbd></td>
<td>Ask Windows to release the TAP adapter lease on shutdown. This option
has no effect now, as it is enabled by default starting with
OpenVPN 2.4.1.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--dhcp-renew</span></kbd></td>
<td>Ask Windows to renew the TAP adapter lease on startup. This option is
normally unnecessary, as Windows automatically triggers a DHCP
renegotiation on the TAP adapter when it comes up, however if you set
the TAP-Win32 adapter Media Status property to "Always Connected", you
may need this flag.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ip-win32 <var>method</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">When using <tt class="docutils literal"><span class="pre">--ifconfig</span></tt> on Windows, set the TAP-Win32 adapter IP
address and netmask using <tt class="docutils literal">method</tt>. Don't use this option unless you
are also using <tt class="docutils literal"><span class="pre">--ifconfig</span></tt>.</p>
<dl class="last docutils">
<dt><code>manual</code></dt>
<dd>Don't set the IP address or netmask automatically. Instead
output a message to the console telling the user to configure the
adapter manually and indicating the IP/netmask which OpenVPN
expects the adapter to be set to.</dd>
<dt><code>dynamic [offset] [lease-time]</code></dt>
<dd><p class="first">Automatically set the IP address and netmask by replying to DHCP
query messages generated by the kernel. This mode is probably the
"cleanest" solution for setting the TCP/IP properties since it
uses the well-known DHCP protocol. There are, however, two
prerequisites for using this mode:</p>
<ol class="arabic simple">
<li>The TCP/IP properties for the TAP-Win32 adapter must be set
to "Obtain an IP address automatically", and</li>
<li>OpenVPN needs to claim an IP address in the subnet for use
as the virtual DHCP server address.</li>
</ol>
<p>By default in <tt class="docutils literal"><span class="pre">--dev</span> tap</tt> mode, OpenVPN will take the normally
unused first address in the subnet. For example, if your subnet is
<code>192.168.4.0 netmask 255.255.255.0</code>, then OpenVPN will take
the IP address <code>192.168.4.0</code> to use as the virtual DHCP
server address. In <tt class="docutils literal"><span class="pre">--dev</span> tun</tt> mode, OpenVPN will cause the DHCP
server to masquerade as if it were coming from the remote endpoint.</p>
<p>The optional offset parameter is an integer which is > <code>-256</code>
and < <code>256</code> and which defaults to 0. If offset is positive,
the DHCP server will masquerade as the IP address at network
address + offset. If offset is negative, the DHCP server will
masquerade as the IP address at broadcast address + offset.</p>
<p>The Windows <code>ipconfig /all</code> command can be used to show what
Windows thinks the DHCP server address is. OpenVPN will "claim"
this address, so make sure to use a free address. Having said that,
different OpenVPN instantiations, including different ends of
the same connection, can share the same virtual DHCP server
address.</p>
<p class="last">The <tt class="docutils literal"><span class="pre">lease-time</span></tt> parameter controls the lease time of the DHCP
assignment given to the TAP-Win32 adapter, and is denoted in
seconds. Normally a very long lease time is preferred because it
prevents routes involving the TAP-Win32 adapter from being lost
when the system goes to sleep. The default lease time is one year.</p>
</dd>
<dt><code>netsh</code></dt>
<dd>Automatically set the IP address and netmask using the Windows
command-line "netsh" command. This method appears to work correctly
on Windows XP but not Windows 2000.</dd>
<dt><code>ipapi</code></dt>
<dd>Automatically set the IP address and netmask using the Windows IP
Helper API. This approach does not have ideal semantics, though
testing has indicated that it works okay in practice. If you use
this option, it is best to leave the TCP/IP properties for the
TAP-Win32 adapter in their default state, i.e. "Obtain an IP
address automatically."</dd>
<dt><code>adaptive</code> (Default)</dt>
<dd><p class="first">Try <code>dynamic</code> method initially and fail over to <code>netsh</code>
if the DHCP negotiation with the TAP-Win32 adapter does not succeed
in 20 seconds. Such failures have been known to occur when certain
third-party firewall packages installed on the client machine block
the DHCP negotiation used by the TAP-Win32 adapter. Note that if
the <code>netsh</code> failover occurs, the TAP-Win32 adapter TCP/IP
properties will be reset from DHCP to static, and this will cause
future OpenVPN startups using the <code>adaptive</code> mode to use
<code>netsh</code> immediately, rather than trying <code>dynamic</code> first.</p>
<p class="last">To "unstick" the <code>adaptive</code> mode from using <code>netsh</code>,
run OpenVPN at least once using the <code>dynamic</code> mode to restore
the TAP-Win32 adapter TCP/IP properties to a DHCP configuration.</p>
</dd>
</dl>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--pause-exit</span></kbd></td>
<td>Put up a "press any key to continue" message on the console prior to
OpenVPN program exit. This option is automatically used by the Windows
explorer when OpenVPN is run on a configuration file using the
right-click explorer menu.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--register-dns</span></kbd></td>
<td>Run <code>ipconfig /flushdns</code> and <code>ipconfig /registerdns</code> on
connection initiation. This is known to kick Windows into recognizing
pushed DNS servers.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--route-method <var>m</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Which method <tt class="docutils literal">m</tt> to use for adding routes on Windows?</p>
<dl class="last docutils">
<dt><code>adaptive</code> (default)</dt>
<dd>Try IP helper API first. If that fails, fall back to the route.exe
shell command.</dd>
<dt><code>ipapi</code></dt>
<dd>Use IP helper API.</dd>
<dt><code>exe</code></dt>
<dd>Call the route.exe shell command.</dd>
</dl>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--service <var>args</var></span></kbd></td>
<td><p class="first">Should be used when OpenVPN is being automatically executed by another
program in such a context that no interaction with the user via display
or keyboard is possible.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
service exit-event [0|1]
</pre>
<p>In general, end-users should never need to explicitly use this option,
as it is automatically added by the OpenVPN service wrapper when a given
OpenVPN configuration is being run as a service.</p>
<p><tt class="docutils literal"><span class="pre">exit-event</span></tt> is the name of a Windows global event object, and OpenVPN
will continuously monitor the state of this event object and exit when
it becomes signaled.</p>
<p>The second parameter indicates the initial state of <tt class="docutils literal"><span class="pre">exit-event</span></tt> and
normally defaults to 0.</p>
<p>Multiple OpenVPN processes can be simultaneously executed with the same
<tt class="docutils literal"><span class="pre">exit-event</span></tt> parameter. In any case, the controlling process can
signal <tt class="docutils literal"><span class="pre">exit-event</span></tt>, causing all such OpenVPN processes to exit.</p>
<p class="last">When executing an OpenVPN process using the <tt class="docutils literal"><span class="pre">--service</span></tt> directive,
OpenVPN will probably not have a console window to output status/error
messages, therefore it is useful to use <tt class="docutils literal"><span class="pre">--log</span></tt> or <tt class="docutils literal"><span class="pre">--log-append</span></tt> to
write these messages to a file.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--show-adapters</span></kbd></td>
</tr>
<tr><td> </td><td>(Standalone) Show available TAP-Win32 adapters which can be selected
using the <tt class="docutils literal"><span class="pre">--dev-node</span></tt> option. On non-Windows systems, the
<tt class="docutils literal">ifconfig</tt>(8) command provides similar functionality.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--show-net</span></kbd></td>
<td>(Standalone) Show OpenVPN's view of the system routing table and network
adapter list.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--show-net-up</span></kbd></td>
<td>Output OpenVPN's view of the system routing table and network adapter
list to the syslog or log file after the TUN/TAP adapter has been
brought up and any routes have been added.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--show-valid-subnets</span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">(Standalone) Show valid subnets for <tt class="docutils literal"><span class="pre">--dev</span> tun</tt> emulation. Since the
TAP-Win32 driver exports an ethernet interface to Windows, and since TUN
devices are point-to-point in nature, it is necessary for the TAP-Win32
driver to impose certain constraints on TUN endpoint address selection.</p>
<p class="last">Namely, the point-to-point endpoints used in TUN device emulation must
be the middle two addresses of a /30 subnet (netmask 255.255.255.252).</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--tap-sleep <var>n</var></span></kbd></td>
<td><p class="first">Cause OpenVPN to sleep for <tt class="docutils literal">n</tt> seconds immediately after the TAP-Win32
adapter state is set to "connected".</p>
<p class="last">This option is intended to be used to troubleshoot problems with the
<tt class="docutils literal"><span class="pre">--ifconfig</span></tt> and <tt class="docutils literal"><span class="pre">--ip-win32</span></tt> options, and is used to give the
TAP-Win32 adapter time to come up before Windows IP Helper API
operations are applied to it.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--win-sys <var>path</var></span></kbd></td>
<td><p class="first">Set the Windows system directory pathname to use when looking for system
executables such as <tt class="docutils literal">route.exe</tt> and <tt class="docutils literal">netsh.exe</tt>. By default, if this
directive is not specified, OpenVPN will use the SystemRoot environment
variable.</p>
<p class="last">This option has changed behaviour since OpenVPN 2.3. Earlier you had to
define <tt class="docutils literal"><span class="pre">--win-sys</span> env</tt> to use the SystemRoot environment variable,
otherwise it defaulted to <code>C:\\WINDOWS</code>. It is not needed to use
the <tt class="docutils literal">env</tt> keyword any more, and it will just be ignored. A warning is
logged when this is found in the configuration file.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--windows-driver <var>drv</var></span></kbd></td>
</tr>
<tr><td> </td><td>Specifies which tun driver to use. Values are <code>ovpn-dco</code> (default),
<code>tap-windows6</code> and <code>wintun</code>. <code>ovpn-dco</code> and <code>wintun</code>
require <tt class="docutils literal"><span class="pre">--dev</span> tun</tt>. <code>wintun</code> also requires OpenVPN process to run
elevated, or be invoked using the Interactive Service.</td></tr>
</tbody>
</table>
</div>
<div class="section" id="standalone-debug-options">
<h2>Standalone Debug Options</h2>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--show-gateway <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">(Standalone) Show current IPv4 and IPv6 default gateway and interface
towards the gateway (if the protocol in question is enabled).</p>
<p>Valid syntax:</p>
<pre class="literal-block">
--show-gateway
--show-gateway IPv6-target
</pre>
<p class="last">For IPv6 this queries the route towards ::/128, or the specified IPv6
target address if passed as argument.
For IPv4 on Linux, Windows, MacOS and BSD it looks for a 0.0.0.0/0 route.
If there are more specific routes, the result will not always be matching
the route of the IPv4 packets to the VPN gateway.</p>
</td></tr>
</tbody>
</table>
</div>
<div class="section" id="advanced-expert-options">
<h2>Advanced Expert Options</h2>
<p>These are options only required when special tweaking is needed, often
used when debugging or testing out special usage scenarios.</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--hash-size <var>args</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Set the size of the real address hash table to <tt class="docutils literal">r</tt> and the virtual
address table to <tt class="docutils literal">v</tt>.</p>
<p>Valid syntax:</p>
<pre class="literal-block">
hash-size r v
</pre>
<p class="last">By default, both tables are sized at 256 buckets.</p>
</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--bcast-buffers <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td>Allocate <tt class="docutils literal">n</tt> buffers for broadcast datagrams (default <code>256</code>).</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--persist-local-ip</span></kbd></td>
</tr>
<tr><td> </td><td>Preserve initially resolved local IP address and port number across
<tt class="docutils literal">SIGUSR1</tt> or <tt class="docutils literal"><span class="pre">--ping-restart</span></tt> restarts.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--persist-remote-ip</span></kbd></td>
</tr>
<tr><td> </td><td>Preserve most recently authenticated remote IP address and port number
across <code>SIGUSR1</code> or <tt class="docutils literal"><span class="pre">--ping-restart</span></tt> restarts.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--rcvbuf <var>size</var></span></kbd></td>
<td>Set the TCP/UDP socket receive buffer size. Defaults to operating system
default.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--shaper <var>n</var></span></kbd></td>
<td><p class="first">Limit bandwidth of outgoing tunnel data to <tt class="docutils literal">n</tt> bytes per second on the
TCP/UDP port. Note that this will only work if mode is set to
<code>p2p</code>. If you want to limit the bandwidth in both directions, use
this option on both peers.</p>
<p>OpenVPN uses the following algorithm to implement traffic shaping: Given
a shaper rate of <tt class="docutils literal">n</tt> bytes per second, after a datagram write of <tt class="docutils literal">b</tt>
bytes is queued on the TCP/UDP port, wait a minimum of <tt class="docutils literal">(b / n)</tt>
seconds before queuing the next write.</p>
<p>It should be noted that OpenVPN supports multiple tunnels between the
same two peers, allowing you to construct full-speed and reduced
bandwidth tunnels at the same time, routing low-priority data such as
off-site backups over the reduced bandwidth tunnel, and other data over
the full-speed tunnel.</p>
<p>Also note that for low bandwidth tunnels (under 1000 bytes per second),
you should probably use lower MTU values as well (see above), otherwise
the packet latency will grow so large as to trigger timeouts in the TLS
layer and TCP connections running over the tunnel.</p>
<p class="last">OpenVPN allows <tt class="docutils literal">n</tt> to be between 100 bytes/sec and 100 Mbytes/sec.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--sndbuf <var>size</var></span></kbd></td>
<td>Set the TCP/UDP socket send buffer size. Defaults to operating system
default.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--tcp-queue-limit <var>n</var></span></kbd></td>
</tr>
<tr><td> </td><td><p class="first">Maximum number of output packets queued before TCP (default <code>64</code>).</p>
<p class="last">When OpenVPN is tunneling data from a TUN/TAP device to a remote client
over a TCP connection, it is possible that the TUN/TAP device might
produce data at a faster rate than the TCP connection can support. When
the number of output packets queued before sending to the TCP socket
reaches this limit for a given client connection, OpenVPN will start to
drop outgoing packets directed at this client.</p>
</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--txqueuelen <var>n</var></span></kbd></td>
<td><em>(Linux only)</em> Set the TX queue length on the TUN/TAP interface.
Currently defaults to operating system default.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--disable-dco</span></kbd></td>
<td><p class="first">Disables the opportunistic use of data channel offloading if available.
Without this option, OpenVPN will opportunistically use DCO mode if
the config options and the running kernel supports using DCO.</p>
<p>Data channel offload currently requires data-ciphers to only contain
AEAD ciphers (AES-GCM and Chacha20-Poly1305) and Linux with the
ovpn-dco module.</p>
<p>Note that some options have no effect or cannot be used when DCO mode
is enabled.</p>
<p class="last">On platforms that do not support DCO <tt class="docutils literal"><span class="pre">disable-dco</span></tt> has no effect.</p>
</td></tr>
</tbody>
</table>
</div>
</div>
<div class="section" id="unsupported-options">
<h1>UNSUPPORTED OPTIONS</h1>
<p>Options listed in this section have been removed from OpenVPN and are no
longer supported</p>
<table class="docutils option-list" frame="void" rules="none">
<col class="option" />
<col class="description" />
<tbody valign="top">
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--client-cert-not-required</span></kbd></td>
</tr>
<tr><td> </td><td>Removed in OpenVPN 2.5. This should be replaxed with
<tt class="docutils literal"><span class="pre">--verify-client-cert</span> none</tt>.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--ifconfig-pool-linear</span></kbd></td>
</tr>
<tr><td> </td><td>Removed in OpenVPN 2.5. This should be replaced with <tt class="docutils literal"><span class="pre">--topology</span> p2p</tt>.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--key-method</span></kbd></td>
<td>Removed in OpenVPN 2.5. This option should not be used, as using the old
<tt class="docutils literal"><span class="pre">key-method</span></tt> weakens the VPN tunnel security. The old <tt class="docutils literal"><span class="pre">key-method</span></tt>
was also only needed when the remote side was older than OpenVPN 2.0.</td></tr>
<tr><td class="option-group" colspan="2">
<kbd><span class="option">--management-client-pf</span></kbd></td>
</tr>
<tr><td> </td><td>Removed in OpenVPN 2.6. The built-in packet filtering (pf) functionality
has been removed.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--ncp-disable</span></kbd></td>
<td>Removed in OpenVPN 2.6. This option mainly served a role as debug option
when NCP was first introduced. It should no longer be necessary.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--no-iv</span></kbd></td>
<td>Removed in OpenVPN 2.5. This option should not be used as it weakens the
VPN tunnel security. This has been a NOOP option since OpenVPN 2.4.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--no-replay</span></kbd></td>
<td>Removed in OpenVPN 2.5. This option should not be used as it weakens the
VPN tunnel security.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--ns-cert-type</span></kbd></td>
<td>Removed in OpenVPN 2.5. The <tt class="docutils literal">nsCertType</tt> field is no longer supported
in recent SSL/TLS libraries. If your certificates does not include <em>key
usage</em> and <em>extended key usage</em> fields, they must be upgraded and the
<tt class="docutils literal"><span class="pre">--remote-cert-tls</span></tt> option should be used instead.</td></tr>
<tr><td class="option-group">
<kbd><span class="option">--prng</span></kbd></td>
<td>Removed in OpenVPN 2.6. We now always use the PRNG of the SSL library.</td></tr>
</tbody>
</table>
</div>
<div class="section" id="connection-profiles">
<h1>CONNECTION PROFILES</h1>
<p>Client configuration files may contain multiple remote servers which
it will attempt to connect against. But there are some configuration
options which are related to specific <tt class="docutils literal"><span class="pre">--remote</span></tt> options. For these
use cases, connection profiles are the solution.</p>
<p>By enacpulating the <tt class="docutils literal"><span class="pre">--remote</span></tt> option and related options within
<tt class="docutils literal"><connection></tt> and <tt class="docutils literal"></connection></tt>, these options are handled as a
group.</p>
<p>An OpenVPN client will try each connection profile sequentially until it
achieves a successful connection.</p>
<p><tt class="docutils literal"><span class="pre">--remote-random</span></tt> can be used to initially "scramble" the connection
list.</p>
<p>Here is an example of connection profile usage:</p>
<pre class="literal-block">
client
dev tun
<connection>
remote 198.19.34.56 1194 udp
</connection>
<connection>
remote 198.19.34.56 443 tcp
</connection>
<connection>
remote 198.19.34.56 443 tcp
http-proxy 192.168.0.8 8080
</connection>
<connection>
remote 198.19.36.99 443 tcp
http-proxy 192.168.0.8 8080
</connection>
persist-key
persist-tun
pkcs12 client.p12
remote-cert-tls server
verb 3
</pre>
<p>First we try to connect to a server at 198.19.34.56:1194 using UDP. If
that fails, we then try to connect to 198.19.34.56:443 using TCP. If
that also fails, then try connecting through an HTTP proxy at
192.168.0.8:8080 to 198.19.34.56:443 using TCP. Finally, try to connect
through the same proxy to a server at 198.19.36.99:443 using TCP.</p>
<p>The following OpenVPN options may be used inside of a <tt class="docutils literal"><connection></tt>
block:</p>
<p><tt class="docutils literal">bind</tt>, <tt class="docutils literal"><span class="pre">connect-retry</span></tt>, <tt class="docutils literal"><span class="pre">connect-retry-max</span></tt>, <tt class="docutils literal"><span class="pre">connect-timeout</span></tt>,
<tt class="docutils literal"><span class="pre">explicit-exit-notify</span></tt>, <tt class="docutils literal">float</tt>, <tt class="docutils literal">fragment</tt>, <tt class="docutils literal"><span class="pre">http-proxy</span></tt>,
<tt class="docutils literal"><span class="pre">http-proxy-option</span></tt>, <tt class="docutils literal"><span class="pre">key-direction</span></tt>, <tt class="docutils literal"><span class="pre">link-mtu</span></tt>, <tt class="docutils literal">local</tt>,
<tt class="docutils literal">lport</tt>, <tt class="docutils literal">mssfix</tt>, <tt class="docutils literal"><span class="pre">mtu-disc</span></tt>, <tt class="docutils literal">nobind</tt>, <tt class="docutils literal">port</tt>, <tt class="docutils literal">proto</tt>,
<tt class="docutils literal">remote</tt>, <tt class="docutils literal">rport</tt>, <tt class="docutils literal"><span class="pre">socks-proxy</span></tt>, <tt class="docutils literal"><span class="pre">tls-auth</span></tt>, <tt class="docutils literal"><span class="pre">tls-crypt</span></tt>,
<tt class="docutils literal"><span class="pre">tun-mtu</span> and</tt>, <tt class="docutils literal"><span class="pre">tun-mtu-extra</span></tt>.</p>
<p>A defaulting mechanism exists for specifying options to apply to all
<tt class="docutils literal"><connection></tt> profiles. If any of the above options (with the
exception of <tt class="docutils literal">remote</tt> ) appear outside of a <tt class="docutils literal"><connection></tt> block,
but in a configuration file which has one or more <tt class="docutils literal"><connection></tt>
blocks, the option setting will be used as a default for
<tt class="docutils literal"><connection></tt> blocks which follow it in the configuration file.</p>
<p>For example, suppose the <tt class="docutils literal">nobind</tt> option were placed in the sample
configuration file above, near the top of the file, before the first
<tt class="docutils literal"><connection></tt> block. The effect would be as if <tt class="docutils literal">nobind</tt> were
declared in all <tt class="docutils literal"><connection></tt> blocks below it.</p>
</div>
<div class="section" id="inline-file-support">
<h1>INLINE FILE SUPPORT</h1>
<p>OpenVPN allows including files in the main configuration for the <tt class="docutils literal"><span class="pre">--ca</span></tt>,
<tt class="docutils literal"><span class="pre">--cert</span></tt>, <tt class="docutils literal"><span class="pre">--dh</span></tt>, <tt class="docutils literal"><span class="pre">--extra-certs</span></tt>, <tt class="docutils literal"><span class="pre">--key</span></tt>, <tt class="docutils literal"><span class="pre">--pkcs12</span></tt>,
<tt class="docutils literal"><span class="pre">--secret</span></tt>, <tt class="docutils literal"><span class="pre">--crl-verify</span></tt>, <tt class="docutils literal"><span class="pre">--http-proxy-user-pass</span></tt>, <tt class="docutils literal"><span class="pre">--tls-auth</span></tt>,
<tt class="docutils literal"><span class="pre">--auth-gen-token-secret</span></tt>, <tt class="docutils literal"><span class="pre">--peer-fingerprint</span></tt>, <tt class="docutils literal"><span class="pre">--tls-crypt</span></tt>,
<tt class="docutils literal"><span class="pre">--tls-crypt-v2</span></tt> and <tt class="docutils literal"><span class="pre">--verify-hash</span></tt> options.</p>
<p>Each inline file started by the line <tt class="docutils literal"><option></tt> and ended by the line
<tt class="docutils literal"></option></tt></p>
<p>Here is an example of an inline file usage</p>
<pre class="literal-block">
<cert>
-----BEGIN CERTIFICATE-----
[...]
-----END CERTIFICATE-----
</cert>
</pre>
<p>When using the inline file feature with <tt class="docutils literal"><span class="pre">--pkcs12</span></tt> the inline file has
to be base64 encoded. Encoding of a .p12 file into base64 can be done
for example with OpenSSL by running <code>openssl base64 -in input.p12</code></p>
</div>
<div class="section" id="signals">
<h1>SIGNALS</h1>
<dl class="docutils">
<dt><code>SIGHUP</code></dt>
<dd>Cause OpenVPN to close all TUN/TAP and network connections, restart,
re-read the configuration file (if any), and reopen TUN/TAP and network
connections.</dd>
<dt><code>SIGUSR1</code></dt>
<dd><p class="first">Like <code>SIGHUP`</code>, except don't re-read configuration file, and
possibly don't close and reopen TUN/TAP device, re-read key files,
preserve local IP address/port, or preserve most recently authenticated
remote IP address/port based on <tt class="docutils literal"><span class="pre">--persist-tun</span></tt>, <tt class="docutils literal"><span class="pre">--persist-key</span></tt>,
<tt class="docutils literal"><span class="pre">--persist-local-ip</span></tt> and <tt class="docutils literal"><span class="pre">--persist-remote-ip</span></tt> options respectively
(see above).</p>
<p>This signal may also be internally generated by a timeout condition,
governed by the <tt class="docutils literal"><span class="pre">--ping-restart</span></tt> option.</p>
<p class="last">This signal, when combined with <tt class="docutils literal"><span class="pre">--persist-remote-ip</span></tt>, may be sent
when the underlying parameters of the host's network interface change
such as when the host is a DHCP client and is assigned a new IP address.
See <tt class="docutils literal"><span class="pre">--ipchange</span></tt> for more information.</p>
</dd>
<dt><code>SIGUSR2</code></dt>
<dd>Causes OpenVPN to display its current statistics (to the syslog file if
<tt class="docutils literal"><span class="pre">--daemon</span></tt> is used, or stdout otherwise).</dd>
<dt><code>SIGINT</code>, <code>SIGTERM</code></dt>
<dd>Causes OpenVPN to exit gracefully.</dd>
</dl>
</div>
<div class="section" id="faq">
<h1>FAQ</h1>
<p><a class="reference external" href="https://community.openvpn.net/openvpn/wiki/FAQ">https://community.openvpn.net/openvpn/wiki/FAQ</a></p>
</div>
<div class="section" id="howto">
<h1>HOWTO</h1>
<p>The manual <tt class="docutils literal"><span class="pre">openvpn-examples</span></tt>(5) gives some examples, especially for
small setups.</p>
<p>For a more comprehensive guide to setting up OpenVPN in a production
setting, see the OpenVPN HOWTO at
<a class="reference external" href="https://openvpn.net/community-resources/how-to/">https://openvpn.net/community-resources/how-to/</a></p>
</div>
<div class="section" id="protocol">
<h1>PROTOCOL</h1>
<p>An ongoing effort to document the OpenVPN protocol can be found under
<a class="reference external" href="https://github.com/openvpn/openvpn-rfc">https://github.com/openvpn/openvpn-rfc</a></p>
</div>
<div class="section" id="web">
<h1>WEB</h1>
<p>OpenVPN's web site is at <a class="reference external" href="https://community.openvpn.net/">https://community.openvpn.net/</a></p>
<p>Go here to download the latest version of OpenVPN, subscribe to the
mailing lists, read the mailing list archives, or browse the Git
repository.</p>
</div>
<div class="section" id="bugs">
<h1>BUGS</h1>
<p>Report all bugs to the OpenVPN team <a class="reference external" href="mailto:info@openvpn.net">info@openvpn.net</a></p>
</div>
<div class="section" id="see-also">
<h1>SEE ALSO</h1>
<p><tt class="docutils literal"><span class="pre">openvpn-examples</span></tt>(5),
<tt class="docutils literal">dhcpcd</tt>(8),
<tt class="docutils literal">ifconfig</tt>(8),
<tt class="docutils literal">openssl</tt>(1),
<tt class="docutils literal">route</tt>(8),
<tt class="docutils literal">scp</tt>(1)
<tt class="docutils literal">ssh</tt>(1)</p>
</div>
<div class="section" id="notes">
<h1>NOTES</h1>
<p>This product includes software developed by the OpenSSL Project
(<a class="reference external" href="https://www.openssl.org/">https://www.openssl.org/</a>)</p>
<p>For more information on the TLS protocol, see
<a class="reference external" href="http://www.ietf.org/rfc/rfc2246.txt">http://www.ietf.org/rfc/rfc2246.txt</a></p>
<p>For more information on the LZO real-time compression library see
<a class="reference external" href="https://www.oberhumer.com/opensource/lzo/">https://www.oberhumer.com/opensource/lzo/</a></p>
</div>
<div class="section" id="copyright">
<h1>COPYRIGHT</h1>
<p>Copyright (C) 2002-2020 OpenVPN Inc This program is free software; you
can redistribute it and/or modify it under the terms of the GNU General
Public License version 2 as published by the Free Software Foundation.</p>
</div>
<div class="section" id="authors">
<h1>AUTHORS</h1>
<p>James Yonan <a class="reference external" href="mailto:james@openvpn.net">james@openvpn.net</a></p>
</div>
</div>
</body>
</html>
Mini Shell