<?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>18.2. Creating a Database Cluster</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="postgres-user.html" title="18.1. The PostgreSQL User Account" /><link rel="next" href="server-start.html" title="18.3. Starting the Database Server" /></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">18.2. Creating a Database Cluster</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="postgres-user.html" title="18.1. The PostgreSQL User Account">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="runtime.html" title="Chapter 18. Server Setup and Operation">Up</a></td><th width="60%" align="center">Chapter 18. Server Setup and Operation</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 11.5 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="server-start.html" title="18.3. Starting the Database Server">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="CREATING-CLUSTER"><div class="titlepage"><div><div><h2 class="title" style="clear: both">18.2. Creating a Database Cluster</h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="creating-cluster.html#CREATING-CLUSTER-MOUNT-POINTS">18.2.1. Use of Secondary File Systems</a></span></dt><dt><span class="sect2"><a href="creating-cluster.html#CREATING-CLUSTER-NFS">18.2.2. Use of Network File Systems</a></span></dt></dl></div><a id="id-1.6.5.4.2" class="indexterm"></a><a id="id-1.6.5.4.3" class="indexterm"></a><p>
Before you can do anything, you must initialize a database storage
area on disk. We call this a <em class="firstterm">database cluster</em>.
(The <acronym class="acronym">SQL</acronym> standard uses the term catalog cluster.) A
database cluster is a collection of databases that is managed by a
single instance of a running database server. After initialization, a
database cluster will contain a database named <code class="literal">postgres</code>,
which is meant as a default database for use by utilities, users and third
party applications. The database server itself does not require the
<code class="literal">postgres</code> database to exist, but many external utility
programs assume it exists. Another database created within each cluster
during initialization is called
<code class="literal">template1</code>. As the name suggests, this will be used
as a template for subsequently created databases; it should not be
used for actual work. (See <a class="xref" href="managing-databases.html" title="Chapter 22. Managing Databases">Chapter 22</a> for
information about creating new databases within a cluster.)
</p><p>
In file system terms, a database cluster is a single directory
under which all data will be stored. We call this the <em class="firstterm">data
directory</em> or <em class="firstterm">data area</em>. It is
completely up to you where you choose to store your data. There is no
default, although locations such as
<code class="filename">/usr/local/pgsql/data</code> or
<code class="filename">/var/lib/pgsql/data</code> are popular. To initialize a
database cluster, use the command <a class="xref" href="app-initdb.html" title="initdb"><span class="refentrytitle">initdb</span></a>,<a id="id-1.6.5.4.5.6" class="indexterm"></a> which is
installed with <span class="productname">PostgreSQL</span>. The desired
file system location of your database cluster is indicated by the
<code class="option">-D</code> option, for example:
</p><pre class="screen">
<code class="prompt">$</code> <strong class="userinput"><code>initdb -D /usr/local/pgsql/data</code></strong>
</pre><p>
Note that you must execute this command while logged into the
<span class="productname">PostgreSQL</span> user account, which is
described in the previous section.
</p><div class="tip"><h3 class="title">Tip</h3><p>
As an alternative to the <code class="option">-D</code> option, you can set
the environment variable <code class="envar">PGDATA</code>.
<a id="id-1.6.5.4.6.1.3" class="indexterm"></a>
</p></div><p>
Alternatively, you can run <code class="command">initdb</code> via
the <a class="xref" href="app-pg-ctl.html" title="pg_ctl"><span class="refentrytitle"><span class="application">pg_ctl</span></span></a>
program<a id="id-1.6.5.4.7.3" class="indexterm"></a> like so:
</p><pre class="screen">
<code class="prompt">$</code> <strong class="userinput"><code>pg_ctl -D /usr/local/pgsql/data initdb</code></strong>
</pre><p>
This may be more intuitive if you are
using <code class="command">pg_ctl</code> for starting and stopping the
server (see <a class="xref" href="server-start.html" title="18.3. Starting the Database Server">Section 18.3</a>), so
that <code class="command">pg_ctl</code> would be the sole command you use
for managing the database server instance.
</p><p>
<code class="command">initdb</code> will attempt to create the directory you
specify if it does not already exist. Of course, this will fail if
<code class="command">initdb</code> does not have permissions to write in the
parent directory. It's generally recommendable that the
<span class="productname">PostgreSQL</span> user own not just the data
directory but its parent directory as well, so that this should not
be a problem. If the desired parent directory doesn't exist either,
you will need to create it first, using root privileges if the
grandparent directory isn't writable. So the process might look
like this:
</p><pre class="screen">
root# <strong class="userinput"><code>mkdir /usr/local/pgsql</code></strong>
root# <strong class="userinput"><code>chown postgres /usr/local/pgsql</code></strong>
root# <strong class="userinput"><code>su postgres</code></strong>
postgres$ <strong class="userinput"><code>initdb -D /usr/local/pgsql/data</code></strong>
</pre><p>
</p><p>
<code class="command">initdb</code> will refuse to run if the data directory
exists and already contains files; this is to prevent accidentally
overwriting an existing installation.
</p><p>
Because the data directory contains all the data stored in the
database, it is essential that it be secured from unauthorized
access. <code class="command">initdb</code> therefore revokes access
permissions from everyone but the
<span class="productname">PostgreSQL</span> user, and optionally, group.
Group access, when enabled, is read-only. This allows an unprivileged
user in the same group as the cluster owner to take a backup of the
cluster data or perform other operations that only require read access.
</p><p>
Note that enabling or disabling group access on an existing cluster requires
the cluster to be shut down and the appropriate mode to be set on all
directories and files before restarting
<span class="productname">PostgreSQL</span>. Otherwise, a mix of modes might
exist in the data directory. For clusters that allow access only by the
owner, the appropriate modes are <code class="literal">0700</code> for directories
and <code class="literal">0600</code> for files. For clusters that also allow
reads by the group, the appropriate modes are <code class="literal">0750</code>
for directories and <code class="literal">0640</code> for files.
</p><p>
However, while the directory contents are secure, the default
client authentication setup allows any local user to connect to the
database and even become the database superuser. If you do not
trust other local users, we recommend you use one of
<code class="command">initdb</code>'s <code class="option">-W</code>, <code class="option">--pwprompt</code>
or <code class="option">--pwfile</code> options to assign a password to the
database superuser.<a id="id-1.6.5.4.12.5" class="indexterm"></a>
Also, specify <code class="option">-A md5</code> or
<code class="option">-A password</code> so that the default <code class="literal">trust</code> authentication
mode is not used; or modify the generated <code class="filename">pg_hba.conf</code>
file after running <code class="command">initdb</code>, but
<span class="emphasis"><em>before</em></span> you start the server for the first time. (Other
reasonable approaches include using <code class="literal">peer</code> authentication
or file system permissions to restrict connections. See <a class="xref" href="client-authentication.html" title="Chapter 20. Client Authentication">Chapter 20</a> for more information.)
</p><p>
<code class="command">initdb</code> also initializes the default
locale<a id="id-1.6.5.4.13.2" class="indexterm"></a> for the database cluster.
Normally, it will just take the locale settings in the environment
and apply them to the initialized database. It is possible to
specify a different locale for the database; more information about
that can be found in <a class="xref" href="locale.html" title="23.1. Locale Support">Section 23.1</a>. The default sort order used
within the particular database cluster is set by
<code class="command">initdb</code>, and while you can create new databases using
different sort order, the order used in the template databases that initdb
creates cannot be changed without dropping and recreating them.
There is also a performance impact for using locales
other than <code class="literal">C</code> or <code class="literal">POSIX</code>. Therefore, it is
important to make this choice correctly the first time.
</p><p>
<code class="command">initdb</code> also sets the default character set encoding
for the database cluster. Normally this should be chosen to match the
locale setting. For details see <a class="xref" href="multibyte.html" title="23.3. Character Set Support">Section 23.3</a>.
</p><p>
Non-<code class="literal">C</code> and non-<code class="literal">POSIX</code> locales rely on the
operating system's collation library for character set ordering.
This controls the ordering of keys stored in indexes. For this reason,
a cluster cannot switch to an incompatible collation library version,
either through snapshot restore, binary streaming replication, a
different operating system, or an operating system upgrade.
</p><div class="sect2" id="CREATING-CLUSTER-MOUNT-POINTS"><div class="titlepage"><div><div><h3 class="title">18.2.1. Use of Secondary File Systems</h3></div></div></div><a id="id-1.6.5.4.16.2" class="indexterm"></a><p>
Many installations create their database clusters on file systems
(volumes) other than the machine's <span class="quote">“<span class="quote">root</span>”</span> volume. If you
choose to do this, it is not advisable to try to use the secondary
volume's topmost directory (mount point) as the data directory.
Best practice is to create a directory within the mount-point
directory that is owned by the <span class="productname">PostgreSQL</span>
user, and then create the data directory within that. This avoids
permissions problems, particularly for operations such
as <span class="application">pg_upgrade</span>, and it also ensures clean failures if
the secondary volume is taken offline.
</p></div><div class="sect2" id="CREATING-CLUSTER-NFS"><div class="titlepage"><div><div><h3 class="title">18.2.2. Use of Network File Systems</h3></div></div></div><a id="id-1.6.5.4.17.2" class="indexterm"></a><a id="id-1.6.5.4.17.3" class="indexterm"></a><a id="id-1.6.5.4.17.4" class="indexterm"></a><p>
Many installations create their database clusters on network file
systems. Sometimes this is done via <acronym class="acronym">NFS</acronym>, or by using a
Network Attached Storage (<acronym class="acronym">NAS</acronym>) device that uses
<acronym class="acronym">NFS</acronym> internally. <span class="productname">PostgreSQL</span> does nothing
special for <acronym class="acronym">NFS</acronym> file systems, meaning it assumes
<acronym class="acronym">NFS</acronym> behaves exactly like locally-connected drives.
If the client or server <acronym class="acronym">NFS</acronym> implementation does not
provide standard file system semantics, this can
cause reliability problems (see <a class="ulink" href="https://www.time-travellers.org/shane/papers/NFS_considered_harmful.html" target="_top">https://www.time-travellers.org/shane/papers/NFS_considered_harmful.html</a>).
Specifically, delayed (asynchronous) writes to the <acronym class="acronym">NFS</acronym>
server can cause data corruption problems. If possible, mount the
<acronym class="acronym">NFS</acronym> file system synchronously (without caching) to avoid
this hazard. Also, soft-mounting the <acronym class="acronym">NFS</acronym> file system is
not recommended.
</p><p>
Storage Area Networks (<acronym class="acronym">SAN</acronym>) typically use communication
protocols other than <acronym class="acronym">NFS</acronym>, and may or may not be subject
to hazards of this sort. It's advisable to consult the vendor's
documentation concerning data consistency guarantees.
<span class="productname">PostgreSQL</span> cannot be more reliable than
the file system it's using.
</p></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="postgres-user.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="runtime.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="server-start.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">18.1. The <span class="productname">PostgreSQL</span> User Account </td><td width="20%" align="center"><a accesskey="h" href="index.html">Home</a></td><td width="40%" align="right" valign="top"> 18.3. Starting the Database Server</td></tr></table></div></body></html>
