| Server IP : 104.21.25.180 / Your IP : 104.23.197.123 Web Server : Apache/2.4.37 System : Linux almalinux.duckdns.org 4.18.0-553.111.1.el8_10.x86_64 #1 SMP Sun Mar 8 20:06:07 EDT 2026 x86_64 User : ricodeal ( 1046) PHP Version : 7.4.33 Disable Function : NONE MySQL : OFF | cURL : ON | WGET : ON | Perl : ON | Python : ON | Sudo : ON | Pkexec : ON Directory : /usr/share/doc/postgresql-docs/html/ |
Upload File : |
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!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"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>20.3. Authentication Methods</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="[email protected]" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="auth-username-maps.html" title="20.2. User Name Maps" /><link rel="next" href="client-authentication-problems.html" title="20.4. Authentication Problems" /></head><body><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">20.3. Authentication Methods</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="auth-username-maps.html" title="20.2. User Name Maps">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="client-authentication.html" title="Chapter 20. Client Authentication">Up</a></td><th width="60%" align="center">Chapter 20. Client Authentication</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 10.23 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="client-authentication-problems.html" title="20.4. Authentication Problems">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="AUTH-METHODS"><div class="titlepage"><div><div><h2 class="title" style="clear: both">20.3. Authentication Methods</h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="auth-methods.html#AUTH-TRUST">20.3.1. Trust Authentication</a></span></dt><dt><span class="sect2"><a href="auth-methods.html#AUTH-PASSWORD">20.3.2. Password Authentication</a></span></dt><dt><span class="sect2"><a href="auth-methods.html#GSSAPI-AUTH">20.3.3. GSSAPI Authentication</a></span></dt><dt><span class="sect2"><a href="auth-methods.html#SSPI-AUTH">20.3.4. SSPI Authentication</a></span></dt><dt><span class="sect2"><a href="auth-methods.html#AUTH-IDENT">20.3.5. Ident Authentication</a></span></dt><dt><span class="sect2"><a href="auth-methods.html#AUTH-PEER">20.3.6. Peer Authentication</a></span></dt><dt><span class="sect2"><a href="auth-methods.html#AUTH-LDAP">20.3.7. LDAP Authentication</a></span></dt><dt><span class="sect2"><a href="auth-methods.html#AUTH-RADIUS">20.3.8. RADIUS Authentication</a></span></dt><dt><span class="sect2"><a href="auth-methods.html#AUTH-CERT">20.3.9. Certificate Authentication</a></span></dt><dt><span class="sect2"><a href="auth-methods.html#AUTH-PAM">20.3.10. PAM Authentication</a></span></dt><dt><span class="sect2"><a href="auth-methods.html#AUTH-BSD">20.3.11. BSD Authentication</a></span></dt></dl></div><p> The following subsections describe the authentication methods in more detail.
</p><div class="sect2" id="AUTH-TRUST"><div class="titlepage"><div><div><h3 class="title">20.3.1. Trust Authentication</h3></div></div></div><p> When <code class="literal">trust</code> authentication is specified,
<span class="productname">PostgreSQL</span> assumes that anyone who can
connect to the server is authorized to access the database with
whatever database user name they specify (even superuser names).
Of course, restrictions made in the <code class="literal">database</code> and
<code class="literal">user</code> columns still apply.
This method should only be used when there is adequate
operating-system-level protection on connections to the server.
</p><p> <code class="literal">trust</code> authentication is appropriate and very
convenient for local connections on a single-user workstation. It
is usually <span class="emphasis"><em>not</em></span> appropriate by itself on a multiuser
machine. However, you might be able to use <code class="literal">trust</code> even
on a multiuser machine, if you restrict access to the server's
Unix-domain socket file using file-system permissions. To do this, set the
<code class="varname">unix_socket_permissions</code> (and possibly
<code class="varname">unix_socket_group</code>) configuration parameters as
described in <a class="xref" href="runtime-config-connection.html" title="19.3. Connections and Authentication">Section 19.3</a>. Or you
could set the <code class="varname">unix_socket_directories</code>
configuration parameter to place the socket file in a suitably
restricted directory.
</p><p> Setting file-system permissions only helps for Unix-socket connections.
Local TCP/IP connections are not restricted by file-system permissions.
Therefore, if you want to use file-system permissions for local security,
remove the <code class="literal">host ... 127.0.0.1 ...</code> line from
<code class="filename">pg_hba.conf</code>, or change it to a
non-<code class="literal">trust</code> authentication method.
</p><p> <code class="literal">trust</code> authentication is only suitable for TCP/IP connections
if you trust every user on every machine that is allowed to connect
to the server by the <code class="filename">pg_hba.conf</code> lines that specify
<code class="literal">trust</code>. It is seldom reasonable to use <code class="literal">trust</code>
for any TCP/IP connections other than those from <span class="systemitem">localhost</span> (127.0.0.1).
</p></div><div class="sect2" id="AUTH-PASSWORD"><div class="titlepage"><div><div><h3 class="title">20.3.2. Password Authentication</h3></div></div></div><a id="id-1.6.7.10.4.2" class="indexterm"></a><a id="id-1.6.7.10.4.3" class="indexterm"></a><a id="id-1.6.7.10.4.4" class="indexterm"></a><p> There are several password-based authentication methods. These methods
operate similarly but differ in how the users' passwords are stored on the
server and how the password provided by a client is sent across the
connection.
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">scram-sha-256</code></span></dt><dd><p> The method <code class="literal">scram-sha-256</code> performs SCRAM-SHA-256
authentication, as described in
<a class="ulink" href="https://tools.ietf.org/html/rfc7677" target="_top">RFC 7677</a>. It
is a challenge-response scheme that prevents password sniffing on
untrusted connections and supports storing passwords on the server in a
cryptographically hashed form that is thought to be secure.
</p><p> This is the most secure of the currently provided methods, but it is
not supported by older client libraries.
</p></dd><dt><span class="term"><code class="literal">md5</code></span></dt><dd><p> The method <code class="literal">md5</code> uses a custom less secure challenge-response
mechanism. It prevents password sniffing and avoids storing passwords
on the server in plain text but provides no protection if an attacker
manages to steal the password hash from the server. Also, the MD5 hash
algorithm is nowadays no longer considered secure against determined
attacks.
</p><p> The <code class="literal">md5</code> method cannot be used with
the <a class="xref" href="runtime-config-connection.html#GUC-DB-USER-NAMESPACE">db_user_namespace</a> feature.
</p><p> To ease transition from the <code class="literal">md5</code> method to the newer
SCRAM method, if <code class="literal">md5</code> is specified as a method
in <code class="filename">pg_hba.conf</code> but the user's password on the
server is encrypted for SCRAM (see below), then SCRAM-based
authentication will automatically be chosen instead.
</p></dd><dt><span class="term"><code class="literal">password</code></span></dt><dd><p> The method <code class="literal">password</code> sends the password in clear-text and is
therefore vulnerable to password <span class="quote">“<span class="quote">sniffing</span>”</span> attacks. It should
always be avoided if possible. If the connection is protected by SSL
encryption then <code class="literal">password</code> can be used safely, though.
(Though SSL certificate authentication might be a better choice if one
is depending on using SSL).
</p></dd></dl></div><p> <span class="productname">PostgreSQL</span> database passwords are
separate from operating system user passwords. The password for
each database user is stored in the <code class="literal">pg_authid</code> system
catalog. Passwords can be managed with the SQL commands
<a class="xref" href="sql-createuser.html" title="CREATE USER"><span class="refentrytitle">CREATE USER</span></a> and
<a class="xref" href="sql-alterrole.html" title="ALTER ROLE"><span class="refentrytitle">ALTER ROLE</span></a>,
e.g., <strong class="userinput"><code>CREATE USER foo WITH PASSWORD 'secret'</code></strong>,
or the <span class="application">psql</span>
command <code class="literal">\password</code>.
If no password has been set up for a user, the stored password
is null and password authentication will always fail for that user.
</p><p> The availability of the different password-based authentication methods
depends on how a user's password on the server is encrypted (or hashed,
more accurately). This is controlled by the configuration
parameter <a class="xref" href="runtime-config-connection.html#GUC-PASSWORD-ENCRYPTION">password_encryption</a> at the time the
password is set. If a password was encrypted using
the <code class="literal">scram-sha-256</code> setting, then it can be used for the
authentication methods <code class="literal">scram-sha-256</code>
and <code class="literal">password</code> (but password transmission will be in
plain text in the latter case). The authentication method
specification <code class="literal">md5</code> will automatically switch to using
the <code class="literal">scram-sha-256</code> method in this case, as explained
above, so it will also work. If a password was encrypted using
the <code class="literal">md5</code> setting, then it can be used only for
the <code class="literal">md5</code> and <code class="literal">password</code> authentication
method specifications (again, with the password transmitted in plain text
in the latter case). (Previous PostgreSQL releases supported storing the
password on the server in plain text. This is no longer possible.) To
check the currently stored password hashes, see the system
catalog <code class="literal">pg_authid</code>.
</p><p> To upgrade an existing installation from <code class="literal">md5</code>
to <code class="literal">scram-sha-256</code>, after having ensured that all client
libraries in use are new enough to support SCRAM,
set <code class="literal">password_encryption = 'scram-sha-256'</code>
in <code class="filename">postgresql.conf</code>, make all users set new passwords,
and change the authentication method specifications
in <code class="filename">pg_hba.conf</code> to <code class="literal">scram-sha-256</code>.
</p></div><div class="sect2" id="GSSAPI-AUTH"><div class="titlepage"><div><div><h3 class="title">20.3.3. GSSAPI Authentication</h3></div></div></div><a id="id-1.6.7.10.5.2" class="indexterm"></a><p> <span class="productname">GSSAPI</span> is an industry-standard protocol
for secure authentication defined in RFC 2743.
<span class="productname">PostgreSQL</span> supports
<span class="productname">GSSAPI</span> with <span class="productname">Kerberos</span>
authentication according to RFC 1964. <span class="productname">GSSAPI</span>
provides automatic authentication (single sign-on) for systems
that support it. The authentication itself is secure, but the
data sent over the database connection will be sent unencrypted unless
<acronym class="acronym">SSL</acronym> is used.
</p><p> GSSAPI support has to be enabled when <span class="productname">PostgreSQL</span> is built;
see <a class="xref" href="installation.html" title="Chapter 16. Installation from Source Code">Chapter 16</a> for more information.
</p><p> When <span class="productname">GSSAPI</span> uses
<span class="productname">Kerberos</span>, it uses a standard principal
in the format
<code class="literal"><em class="replaceable"><code>servicename</code></em>/<em class="replaceable"><code>hostname</code></em>@<em class="replaceable"><code>realm</code></em></code>.
The PostgreSQL server will accept any principal that is included in the keytab used by
the server, but care needs to be taken to specify the correct principal details when
making the connection from the client using the <code class="literal">krbsrvname</code> connection parameter. (See
also <a class="xref" href="libpq-connect.html#LIBPQ-PARAMKEYWORDS" title="33.1.2. Parameter Key Words">Section 33.1.2</a>.) The installation default can be
changed from the default <code class="literal">postgres</code> at build time using
<code class="literal">./configure --with-krb-srvnam=</code><em class="replaceable"><code>whatever</code></em>.
In most environments,
this parameter never needs to be changed.
Some Kerberos implementations might require a different service name,
such as Microsoft Active Directory which requires the service name
to be in upper case (<code class="literal">POSTGRES</code>).
</p><p> <em class="replaceable"><code>hostname</code></em> is the fully qualified host name of the
server machine. The service principal's realm is the preferred realm
of the server machine.
</p><p> Client principals can be mapped to different <span class="productname">PostgreSQL</span>
database user names with <code class="filename">pg_ident.conf</code>. For example,
<code class="literal">pgusername@realm</code> could be mapped to just <code class="literal">pgusername</code>.
Alternatively, you can use the full <code class="literal">username@realm</code> principal as
the role name in <span class="productname">PostgreSQL</span> without any mapping.
</p><p> <span class="productname">PostgreSQL</span> also supports a parameter to strip the realm from
the principal. This method is supported for backwards compatibility and is
strongly discouraged as it is then impossible to distinguish different users
with the same user name but coming from different realms. To enable this,
set <code class="literal">include_realm</code> to 0. For simple single-realm
installations, doing that combined with setting the
<code class="literal">krb_realm</code> parameter (which checks that the principal's realm
matches exactly what is in the <code class="literal">krb_realm</code> parameter)
is still secure; but this is a
less capable approach compared to specifying an explicit mapping in
<code class="filename">pg_ident.conf</code>.
</p><p> Make sure that your server keytab file is readable (and preferably
only readable, not writable) by the <span class="productname">PostgreSQL</span>
server account. (See also <a class="xref" href="postgres-user.html" title="18.1. The PostgreSQL User Account">Section 18.1</a>.) The location
of the key file is specified by the <a class="xref" href="runtime-config-connection.html#GUC-KRB-SERVER-KEYFILE">krb_server_keyfile</a> configuration
parameter. The default is
<code class="filename">/usr/local/pgsql/etc/krb5.keytab</code> (or whatever
directory was specified as <code class="varname">sysconfdir</code> at build time).
For security reasons, it is recommended to use a separate keytab
just for the <span class="productname">PostgreSQL</span> server rather
than opening up permissions on the system keytab file.
</p><p> The keytab file is generated by the Kerberos software; see the
Kerberos documentation for details. The following example is
for MIT-compatible Kerberos 5 implementations:
</p><pre class="screen"><code class="prompt">kadmin% </code><strong class="userinput"><code>ank -randkey postgres/server.my.domain.org</code></strong>
<code class="prompt">kadmin% </code><strong class="userinput"><code>ktadd -k krb5.keytab postgres/server.my.domain.org</code></strong></pre><p>
</p><p> When connecting to the database make sure you have a ticket for a
principal matching the requested database user name. For example, for
database user name <code class="literal">fred</code>, principal
<code class="literal">[email protected]</code> would be able to connect. To also allow
principal <code class="literal">fred/[email protected]</code>, use a user name
map, as described in <a class="xref" href="auth-username-maps.html" title="20.2. User Name Maps">Section 20.2</a>.
</p><p> The following configuration options are supported for <span class="productname">GSSAPI</span>:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">include_realm</code></span></dt><dd><p> If set to 0, the realm name from the authenticated user principal is
stripped off before being passed through the user name mapping
(<a class="xref" href="auth-username-maps.html" title="20.2. User Name Maps">Section 20.2</a>). This is discouraged and is
primarily available for backwards compatibility, as it is not secure
in multi-realm environments unless <code class="literal">krb_realm</code> is
also used. It is recommended to
leave <code class="literal">include_realm</code> set to the default (1) and to
provide an explicit mapping in <code class="filename">pg_ident.conf</code> to convert
principal names to <span class="productname">PostgreSQL</span> user names.
</p></dd><dt><span class="term"><code class="literal">map</code></span></dt><dd><p> Allows for mapping between system and database user names. See
<a class="xref" href="auth-username-maps.html" title="20.2. User Name Maps">Section 20.2</a> for details. For a GSSAPI/Kerberos
principal, such as <code class="literal">[email protected]</code> (or, less
commonly, <code class="literal">username/[email protected]</code>), the
user name used for mapping is
<code class="literal">[email protected]</code> (or
<code class="literal">username/[email protected]</code>, respectively),
unless <code class="literal">include_realm</code> has been set to 0, in which case
<code class="literal">username</code> (or <code class="literal">username/hostbased</code>)
is what is seen as the system user name when mapping.
</p></dd><dt><span class="term"><code class="literal">krb_realm</code></span></dt><dd><p> Sets the realm to match user principal names against. If this parameter
is set, only users of that realm will be accepted. If it is not set,
users of any realm can connect, subject to whatever user name mapping
is done.
</p></dd></dl></div><p>
</p></div><div class="sect2" id="SSPI-AUTH"><div class="titlepage"><div><div><h3 class="title">20.3.4. SSPI Authentication</h3></div></div></div><a id="id-1.6.7.10.6.2" class="indexterm"></a><p> <span class="productname">SSPI</span> is a <span class="productname">Windows</span>
technology for secure authentication with single sign-on.
<span class="productname">PostgreSQL</span> will use SSPI in
<code class="literal">negotiate</code> mode, which will use
<span class="productname">Kerberos</span> when possible and automatically
fall back to <span class="productname">NTLM</span> in other cases.
<span class="productname">SSPI</span> authentication only works when both
server and client are running <span class="productname">Windows</span>,
or, on non-Windows platforms, when <span class="productname">GSSAPI</span>
is available.
</p><p> When using <span class="productname">Kerberos</span> authentication,
<span class="productname">SSPI</span> works the same way
<span class="productname">GSSAPI</span> does; see <a class="xref" href="auth-methods.html#GSSAPI-AUTH" title="20.3.3. GSSAPI Authentication">Section 20.3.3</a>
for details.
</p><p> The following configuration options are supported for <span class="productname">SSPI</span>:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">include_realm</code></span></dt><dd><p> If set to 0, the realm name from the authenticated user principal is
stripped off before being passed through the user name mapping
(<a class="xref" href="auth-username-maps.html" title="20.2. User Name Maps">Section 20.2</a>). This is discouraged and is
primarily available for backwards compatibility, as it is not secure
in multi-realm environments unless <code class="literal">krb_realm</code> is
also used. It is recommended to
leave <code class="literal">include_realm</code> set to the default (1) and to
provide an explicit mapping in <code class="filename">pg_ident.conf</code> to convert
principal names to <span class="productname">PostgreSQL</span> user names.
</p></dd><dt><span class="term"><code class="literal">compat_realm</code></span></dt><dd><p> If set to 1, the domain's SAM-compatible name (also known as the
NetBIOS name) is used for the <code class="literal">include_realm</code>
option. This is the default. If set to 0, the true realm name from
the Kerberos user principal name is used.
</p><p> Do not disable this option unless your server runs under a domain
account (this includes virtual service accounts on a domain member
system) and all clients authenticating through SSPI are also using
domain accounts, or authentication will fail.
</p></dd><dt><span class="term"><code class="literal">upn_username</code></span></dt><dd><p> If this option is enabled along with <code class="literal">compat_realm</code>,
the user name from the Kerberos UPN is used for authentication. If
it is disabled (the default), the SAM-compatible user name is used.
By default, these two names are identical for new user accounts.
</p><p> Note that <span class="application">libpq</span> uses the SAM-compatible name if no
explicit user name is specified. If you use
<span class="application">libpq</span> or a driver based on it, you should
leave this option disabled or explicitly specify user name in the
connection string.
</p></dd><dt><span class="term"><code class="literal">map</code></span></dt><dd><p> Allows for mapping between system and database user names. See
<a class="xref" href="auth-username-maps.html" title="20.2. User Name Maps">Section 20.2</a> for details. For a SSPI/Kerberos
principal, such as <code class="literal">[email protected]</code> (or, less
commonly, <code class="literal">username/[email protected]</code>), the
user name used for mapping is
<code class="literal">[email protected]</code> (or
<code class="literal">username/[email protected]</code>, respectively),
unless <code class="literal">include_realm</code> has been set to 0, in which case
<code class="literal">username</code> (or <code class="literal">username/hostbased</code>)
is what is seen as the system user name when mapping.
</p></dd><dt><span class="term"><code class="literal">krb_realm</code></span></dt><dd><p> Sets the realm to match user principal names against. If this parameter
is set, only users of that realm will be accepted. If it is not set,
users of any realm can connect, subject to whatever user name mapping
is done.
</p></dd></dl></div><p>
</p></div><div class="sect2" id="AUTH-IDENT"><div class="titlepage"><div><div><h3 class="title">20.3.5. Ident Authentication</h3></div></div></div><a id="id-1.6.7.10.7.2" class="indexterm"></a><p> The ident authentication method works by obtaining the client's
operating system user name from an ident server and using it as
the allowed database user name (with an optional user name mapping).
This is only supported on TCP/IP connections.
</p><div class="note"><h3 class="title">Note</h3><p> When ident is specified for a local (non-TCP/IP) connection,
peer authentication (see <a class="xref" href="auth-methods.html#AUTH-PEER" title="20.3.6. Peer Authentication">Section 20.3.6</a>) will be
used instead.
</p></div><p> The following configuration options are supported for <span class="productname">ident</span>:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">map</code></span></dt><dd><p> Allows for mapping between system and database user names. See
<a class="xref" href="auth-username-maps.html" title="20.2. User Name Maps">Section 20.2</a> for details.
</p></dd></dl></div><p>
</p><p> The <span class="quote">“<span class="quote">Identification Protocol</span>”</span> is described in
RFC 1413. Virtually every Unix-like
operating system ships with an ident server that listens on TCP
port 113 by default. The basic functionality of an ident server
is to answer questions like <span class="quote">“<span class="quote">What user initiated the
connection that goes out of your port <em class="replaceable"><code>X</code></em>
and connects to my port <em class="replaceable"><code>Y</code></em>?</span>”</span>.
Since <span class="productname">PostgreSQL</span> knows both <em class="replaceable"><code>X</code></em> and
<em class="replaceable"><code>Y</code></em> when a physical connection is established, it
can interrogate the ident server on the host of the connecting
client and can theoretically determine the operating system user
for any given connection.
</p><p> The drawback of this procedure is that it depends on the integrity
of the client: if the client machine is untrusted or compromised,
an attacker could run just about any program on port 113 and
return any user name they choose. This authentication method is
therefore only appropriate for closed networks where each client
machine is under tight control and where the database and system
administrators operate in close contact. In other words, you must
trust the machine running the ident server.
Heed the warning:
</p><div class="blockquote"><table border="0" class="blockquote" style="width: 100%; cellspacing: 0; cellpadding: 0;" summary="Block quote"><tr><td width="10%" valign="top"> </td><td width="80%" valign="top"><p> The Identification Protocol is not intended as an authorization
or access control protocol.
</p></td><td width="10%" valign="top"> </td></tr><tr><td width="10%" valign="top"> </td><td colspan="2" align="right" valign="top">--<span class="attribution">RFC 1413</span></td></tr></table></div><p>
</p><p> Some ident servers have a nonstandard option that causes the returned
user name to be encrypted, using a key that only the originating
machine's administrator knows. This option <span class="emphasis"><em>must not</em></span> be
used when using the ident server with <span class="productname">PostgreSQL</span>,
since <span class="productname">PostgreSQL</span> does not have any way to decrypt the
returned string to determine the actual user name.
</p></div><div class="sect2" id="AUTH-PEER"><div class="titlepage"><div><div><h3 class="title">20.3.6. Peer Authentication</h3></div></div></div><a id="id-1.6.7.10.8.2" class="indexterm"></a><p> The peer authentication method works by obtaining the client's
operating system user name from the kernel and using it as the
allowed database user name (with optional user name mapping). This
method is only supported on local connections.
</p><p> The following configuration options are supported for <span class="productname">peer</span>:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">map</code></span></dt><dd><p> Allows for mapping between system and database user names. See
<a class="xref" href="auth-username-maps.html" title="20.2. User Name Maps">Section 20.2</a> for details.
</p></dd></dl></div><p>
</p><p> Peer authentication is only available on operating systems providing
the <code class="function">getpeereid()</code> function, the <code class="symbol">SO_PEERCRED</code>
socket parameter, or similar mechanisms. Currently that includes
<span class="systemitem">Linux</span>,
most flavors of <span class="systemitem">BSD</span> including
<span class="systemitem">macOS</span>,
and <span class="systemitem">Solaris</span>.
</p></div><div class="sect2" id="AUTH-LDAP"><div class="titlepage"><div><div><h3 class="title">20.3.7. LDAP Authentication</h3></div></div></div><a id="id-1.6.7.10.9.2" class="indexterm"></a><p> This authentication method operates similarly to
<code class="literal">password</code> except that it uses LDAP
as the password verification method. LDAP is used only to validate
the user name/password pairs. Therefore the user must already
exist in the database before LDAP can be used for
authentication.
</p><p> LDAP authentication can operate in two modes. In the first mode,
which we will call the simple bind mode,
the server will bind to the distinguished name constructed as
<em class="replaceable"><code>prefix</code></em> <em class="replaceable"><code>username</code></em> <em class="replaceable"><code>suffix</code></em>.
Typically, the <em class="replaceable"><code>prefix</code></em> parameter is used to specify
<code class="literal">cn=</code>, or <em class="replaceable"><code>DOMAIN</code></em><code class="literal">\</code> in an Active
Directory environment. <em class="replaceable"><code>suffix</code></em> is used to specify the
remaining part of the DN in a non-Active Directory environment.
</p><p> In the second mode, which we will call the search+bind mode,
the server first binds to the LDAP directory with
a fixed user name and password, specified with <em class="replaceable"><code>ldapbinddn</code></em>
and <em class="replaceable"><code>ldapbindpasswd</code></em>, and performs a search for the user trying
to log in to the database. If no user and password is configured, an
anonymous bind will be attempted to the directory. The search will be
performed over the subtree at <em class="replaceable"><code>ldapbasedn</code></em>, and will try to
do an exact match of the attribute specified in
<em class="replaceable"><code>ldapsearchattribute</code></em>.
Once the user has been found in
this search, the server disconnects and re-binds to the directory as
this user, using the password specified by the client, to verify that the
login is correct. This mode is the same as that used by LDAP authentication
schemes in other software, such as Apache <code class="literal">mod_authnz_ldap</code> and <code class="literal">pam_ldap</code>.
This method allows for significantly more flexibility
in where the user objects are located in the directory, but will cause
two separate connections to the LDAP server to be made.
</p><p> The following configuration options are used in both modes:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">ldapserver</code></span></dt><dd><p> Names or IP addresses of LDAP servers to connect to. Multiple
servers may be specified, separated by spaces.
</p></dd><dt><span class="term"><code class="literal">ldapport</code></span></dt><dd><p> Port number on LDAP server to connect to. If no port is specified,
the LDAP library's default port setting will be used.
</p></dd><dt><span class="term"><code class="literal">ldaptls</code></span></dt><dd><p> Set to 1 to make the connection between PostgreSQL and the
LDAP server use TLS encryption. Note that this only encrypts
the traffic to the LDAP server — the connection to the client
will still be unencrypted unless SSL is used.
</p></dd></dl></div><p>
The following options are used in simple bind mode only:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">ldapprefix</code></span></dt><dd><p> String to prepend to the user name when forming the DN to bind as,
when doing simple bind authentication.
</p></dd><dt><span class="term"><code class="literal">ldapsuffix</code></span></dt><dd><p> String to append to the user name when forming the DN to bind as,
when doing simple bind authentication.
</p></dd></dl></div><p>
The following options are used in search+bind mode only:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">ldapbasedn</code></span></dt><dd><p> Root DN to begin the search for the user in, when doing search+bind
authentication.
</p></dd><dt><span class="term"><code class="literal">ldapbinddn</code></span></dt><dd><p> DN of user to bind to the directory with to perform the search when
doing search+bind authentication.
</p></dd><dt><span class="term"><code class="literal">ldapbindpasswd</code></span></dt><dd><p> Password for user to bind to the directory with to perform the search
when doing search+bind authentication.
</p></dd><dt><span class="term"><code class="literal">ldapsearchattribute</code></span></dt><dd><p> Attribute to match against the user name in the search when doing
search+bind authentication. If no attribute is specified, the
<code class="literal">uid</code> attribute will be used.
</p></dd><dt><span class="term"><code class="literal">ldapurl</code></span></dt><dd><p> An RFC 4516 LDAP URL. This is an alternative way to write some of the
other LDAP options in a more compact and standard form. The format is
</p><pre class="synopsis">ldap://<em class="replaceable"><code>host</code></em>[:<em class="replaceable"><code>port</code></em>]/<em class="replaceable"><code>basedn</code></em>[?[<em class="replaceable"><code>attribute</code></em>][?[<em class="replaceable"><code>scope</code></em>]]]</pre><p>
<em class="replaceable"><code>scope</code></em> must be one
of <code class="literal">base</code>, <code class="literal">one</code>, <code class="literal">sub</code>,
typically the latter. Only one attribute is used, and some other
components of standard LDAP URLs such as filters and extensions are
not supported.
</p><p> For non-anonymous binds, <code class="literal">ldapbinddn</code>
and <code class="literal">ldapbindpasswd</code> must be specified as separate
options.
</p><p> To use encrypted LDAP connections, the <code class="literal">ldaptls</code>
option has to be used in addition to <code class="literal">ldapurl</code>.
The <code class="literal">ldaps</code> URL scheme (direct SSL connection) is not
supported.
</p><p> LDAP URLs are currently only supported with OpenLDAP, not on Windows.
</p></dd></dl></div><p>
</p><p> It is an error to mix configuration options for simple bind with options
for search+bind.
</p><p> Here is an example for a simple-bind LDAP configuration:
</p><pre class="programlisting">host ... ldap ldapserver=ldap.example.net ldapprefix="cn=" ldapsuffix=", dc=example, dc=net"</pre><p>
When a connection to the database server as database
user <code class="literal">someuser</code> is requested, PostgreSQL will attempt to
bind to the LDAP server using the DN <code class="literal">cn=someuser, dc=example,
dc=net</code> and the password provided by the client. If that connection
succeeds, the database access is granted.
</p><p> Here is an example for a search+bind configuration:
</p><pre class="programlisting">host ... ldap ldapserver=ldap.example.net ldapbasedn="dc=example, dc=net" ldapsearchattribute=uid</pre><p>
When a connection to the database server as database
user <code class="literal">someuser</code> is requested, PostgreSQL will attempt to
bind anonymously (since <code class="literal">ldapbinddn</code> was not specified) to
the LDAP server, perform a search for <code class="literal">(uid=someuser)</code>
under the specified base DN. If an entry is found, it will then attempt to
bind using that found information and the password supplied by the client.
If that second connection succeeds, the database access is granted.
</p><p> Here is the same search+bind configuration written as a URL:
</p><pre class="programlisting">host ... ldap ldapurl="ldap://ldap.example.net/dc=example,dc=net?uid?sub"</pre><p>
Some other software that supports authentication against LDAP uses the
same URL format, so it will be easier to share the configuration.
</p><div class="tip"><h3 class="title">Tip</h3><p> Since LDAP often uses commas and spaces to separate the different
parts of a DN, it is often necessary to use double-quoted parameter
values when configuring LDAP options, as shown in the examples.
</p></div></div><div class="sect2" id="AUTH-RADIUS"><div class="titlepage"><div><div><h3 class="title">20.3.8. RADIUS Authentication</h3></div></div></div><a id="id-1.6.7.10.10.2" class="indexterm"></a><p> This authentication method operates similarly to
<code class="literal">password</code> except that it uses RADIUS
as the password verification method. RADIUS is used only to validate
the user name/password pairs. Therefore the user must already
exist in the database before RADIUS can be used for
authentication.
</p><p> When using RADIUS authentication, an Access Request message will be sent
to the configured RADIUS server. This request will be of type
<code class="literal">Authenticate Only</code>, and include parameters for
<code class="literal">user name</code>, <code class="literal">password</code> (encrypted) and
<code class="literal">NAS Identifier</code>. The request will be encrypted using
a secret shared with the server. The RADIUS server will respond to
this request with either <code class="literal">Access Accept</code> or
<code class="literal">Access Reject</code>. There is no support for RADIUS accounting.
</p><p> Multiple RADIUS servers can be specified, in which case they will
be tried sequentially. If a negative response is received from
a server, the authentication will fail. If no response is received,
the next server in the list will be tried. To specify multiple
servers, separate the server names with commas and surround the list
with double quotes. If multiple servers are specified, the other
RADIUS options can also be given as comma-separated lists, to provide
individual values for each server. They can also be specified as
a single value, in which case that value will apply to all servers.
</p><p> The following configuration options are supported for RADIUS:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">radiusservers</code></span></dt><dd><p> The DNS names or IP addresses of the RADIUS servers to connect to.
This parameter is required.
</p></dd><dt><span class="term"><code class="literal">radiussecrets</code></span></dt><dd><p> The shared secrets used when talking securely to the RADIUS
servers. This must have exactly the same value on the PostgreSQL
and RADIUS servers. It is recommended that this be a string of
at least 16 characters. This parameter is required.
</p><div class="note"><h3 class="title">Note</h3><p> The encryption vector used will only be cryptographically
strong if <span class="productname">PostgreSQL</span> is built with support for
<span class="productname">OpenSSL</span>. In other cases, the transmission to the
RADIUS server should only be considered obfuscated, not secured, and
external security measures should be applied if necessary.
</p></div><p>
</p></dd><dt><span class="term"><code class="literal">radiusports</code></span></dt><dd><p> The port numbers to connect to on the RADIUS servers. If no port
is specified, the default RADIUS port (<code class="literal">1812</code>)
will be used.
</p></dd><dt><span class="term"><code class="literal">radiusidentifiers</code></span></dt><dd><p> The strings to be used as <code class="literal">NAS Identifier</code> in the
RADIUS requests. This parameter can be used, for example, to
identify which database cluster the user is attempting to connect
to, which can be useful for policy matching on
the RADIUS server. If no identifier is specified, the default
<code class="literal">postgresql</code> will be used.
</p></dd></dl></div><p>
</p><p> If it is necessary to have a comma or whitespace in a RADIUS parameter
value, that can be done by putting double quotes around the value, but
it is tedious because two layers of double-quoting are now required.
An example of putting whitespace into RADIUS secret strings is:
</p><pre class="programlisting">host ... radius radiusservers="server1,server2" radiussecrets="""secret one"",""secret two"""</pre><p>
</p></div><div class="sect2" id="AUTH-CERT"><div class="titlepage"><div><div><h3 class="title">20.3.9. Certificate Authentication</h3></div></div></div><a id="id-1.6.7.10.11.2" class="indexterm"></a><p> This authentication method uses SSL client certificates to perform
authentication. It is therefore only available for SSL connections.
When using this authentication method, the server will require that
the client provide a valid, trusted certificate. No password prompt
will be sent to the client. The <code class="literal">cn</code> (Common Name)
attribute of the certificate
will be compared to the requested database user name, and if they match
the login will be allowed. User name mapping can be used to allow
<code class="literal">cn</code> to be different from the database user name.
</p><p> The following configuration options are supported for SSL certificate
authentication:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">map</code></span></dt><dd><p> Allows for mapping between system and database user names. See
<a class="xref" href="auth-username-maps.html" title="20.2. User Name Maps">Section 20.2</a> for details.
</p></dd></dl></div><p>
</p><p> In a <code class="filename">pg_hba.conf</code> record specifying certificate
authentication, the authentication option <code class="literal">clientcert</code> is
assumed to be <code class="literal">1</code>, and it cannot be turned off since a client
certificate is necessary for this method. What the <code class="literal">cert</code>
method adds to the basic <code class="literal">clientcert</code> certificate validity test
is a check that the <code class="literal">cn</code> attribute matches the database
user name.
</p></div><div class="sect2" id="AUTH-PAM"><div class="titlepage"><div><div><h3 class="title">20.3.10. PAM Authentication</h3></div></div></div><a id="id-1.6.7.10.12.2" class="indexterm"></a><p> This authentication method operates similarly to
<code class="literal">password</code> except that it uses PAM (Pluggable
Authentication Modules) as the authentication mechanism. The
default PAM service name is <code class="literal">postgresql</code>.
PAM is used only to validate user name/password pairs and optionally the
connected remote host name or IP address. Therefore the user must already
exist in the database before PAM can be used for authentication. For more
information about PAM, please read the
<a class="ulink" href="http://www.kernel.org/pub/linux/libs/pam/" target="_top"> <span class="productname">Linux-PAM</span> Page</a>.
</p><p> The following configuration options are supported for PAM:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">pamservice</code></span></dt><dd><p> PAM service name.
</p></dd><dt><span class="term"><code class="literal">pam_use_hostname</code></span></dt><dd><p> Determines whether the remote IP address or the host name is provided
to PAM modules through the <code class="symbol">PAM_RHOST</code> item. By
default, the IP address is used. Set this option to 1 to use the
resolved host name instead. Host name resolution can lead to login
delays. (Most PAM configurations don't use this information, so it is
only necessary to consider this setting if a PAM configuration was
specifically created to make use of it.)
</p></dd></dl></div><p>
</p><div class="note"><h3 class="title">Note</h3><p> If PAM is set up to read <code class="filename">/etc/shadow</code>, authentication
will fail because the PostgreSQL server is started by a non-root
user. However, this is not an issue when PAM is configured to use
LDAP or other authentication methods.
</p></div></div><div class="sect2" id="AUTH-BSD"><div class="titlepage"><div><div><h3 class="title">20.3.11. BSD Authentication</h3></div></div></div><a id="id-1.6.7.10.13.2" class="indexterm"></a><p> This authentication method operates similarly to
<code class="literal">password</code> except that it uses BSD Authentication
to verify the password. BSD Authentication is used only
to validate user name/password pairs. Therefore the user's role must
already exist in the database before BSD Authentication can be used
for authentication. The BSD Authentication framework is currently
only available on OpenBSD.
</p><p> BSD Authentication in <span class="productname">PostgreSQL</span> uses
the <code class="literal">auth-postgresql</code> login type and authenticates with
the <code class="literal">postgresql</code> login class if that's defined
in <code class="filename">login.conf</code>. By default that login class does not
exist, and <span class="productname">PostgreSQL</span> will use the default login class.
</p><div class="note"><h3 class="title">Note</h3><p> To use BSD Authentication, the PostgreSQL user account (that is, the
operating system user running the server) must first be added to
the <code class="literal">auth</code> group. The <code class="literal">auth</code> group
exists by default on OpenBSD systems.
</p></div></div></div><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navfooter"><hr></hr><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="auth-username-maps.html" title="20.2. User Name Maps">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="client-authentication.html" title="Chapter 20. Client Authentication">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="client-authentication-problems.html" title="20.4. Authentication Problems">Next</a></td></tr><tr><td width="40%" align="left" valign="top">20.2. User Name Maps </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 10.23 Documentation">Home</a></td><td width="40%" align="right" valign="top"> 20.4. Authentication Problems</td></tr></table></div></body></html>