<?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>16.4. Installation Procedure</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="install-getsource.html" title="16.3. Getting The Source" /><link rel="next" href="install-post.html" title="16.5. Post-Installation Setup" /></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">16.4. Installation Procedure</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="install-getsource.html" title="16.3. Getting The Source">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="installation.html" title="Chapter 16. Installation from Source Code">Up</a></td><th width="60%" align="center">Chapter 16. Installation from Source Code</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="install-post.html" title="16.5. Post-Installation Setup">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="INSTALL-PROCEDURE"><div class="titlepage"><div><div><h2 class="title" style="clear: both">16.4. Installation Procedure</h2></div></div></div><div class="procedure"><ol class="procedure" type="1"><li class="step" id="CONFIGURE"><p class="title"><strong>Configuration</strong></p><a id="id-1.6.3.7.2.1.2" class="indexterm"></a><p>
The first step of the installation procedure is to configure the
source tree for your system and choose the options you would like.
This is done by running the <code class="filename">configure</code> script. For a
default installation simply enter:
</p><pre class="screen">
<strong class="userinput"><code>./configure</code></strong>
</pre><p>
This script will run a number of tests to determine values for various
system dependent variables and detect any quirks of your
operating system, and finally will create several files in the
build tree to record what it found. You can also run
<code class="filename">configure</code> in a directory outside the source
tree, if you want to keep the build directory separate. This
procedure is also called a
<a id="id-1.6.3.7.2.1.3.4" class="indexterm"></a><em class="firstterm">VPATH</em>
build. Here's how:
</p><pre class="screen">
<strong class="userinput"><code>mkdir build_dir</code></strong>
<strong class="userinput"><code>cd build_dir</code></strong>
<strong class="userinput"><code>/path/to/source/tree/configure [options go here]</code></strong>
<strong class="userinput"><code>make</code></strong>
</pre><p>
</p><p>
The default configuration will build the server and utilities, as
well as all client applications and interfaces that require only a
C compiler. All files will be installed under
<code class="filename">/usr/local/pgsql</code> by default.
</p><p>
You can customize the build and installation process by supplying one
or more of the following command line options to
<code class="filename">configure</code>:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="option">--prefix=<em class="replaceable"><code>PREFIX</code></em></code></span></dt><dd><p>
Install all files under the directory <em class="replaceable"><code>PREFIX</code></em>
instead of <code class="filename">/usr/local/pgsql</code>. The actual
files will be installed into various subdirectories; no files
will ever be installed directly into the
<em class="replaceable"><code>PREFIX</code></em> directory.
</p><p>
If you have special needs, you can also customize the
individual subdirectories with the following options. However,
if you leave these with their defaults, the installation will be
relocatable, meaning you can move the directory after
installation. (The <code class="literal">man</code> and <code class="literal">doc</code>
locations are not affected by this.)
</p><p>
For relocatable installs, you might want to use
<code class="filename">configure</code>'s <code class="literal">--disable-rpath</code>
option. Also, you will need to tell the operating system how
to find the shared libraries.
</p></dd><dt><span class="term"><code class="option">--exec-prefix=<em class="replaceable"><code>EXEC-PREFIX</code></em></code></span></dt><dd><p>
You can install architecture-dependent files under a
different prefix, <em class="replaceable"><code>EXEC-PREFIX</code></em>, than what
<em class="replaceable"><code>PREFIX</code></em> was set to. This can be useful to
share architecture-independent files between hosts. If you
omit this, then <em class="replaceable"><code>EXEC-PREFIX</code></em> is set equal to
<em class="replaceable"><code>PREFIX</code></em> and both architecture-dependent and
independent files will be installed under the same tree,
which is probably what you want.
</p></dd><dt><span class="term"><code class="option">--bindir=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
Specifies the directory for executable programs. The default
is <code class="filename"><em class="replaceable"><code>EXEC-PREFIX</code></em>/bin</code>, which
normally means <code class="filename">/usr/local/pgsql/bin</code>.
</p></dd><dt><span class="term"><code class="option">--sysconfdir=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
Sets the directory for various configuration files,
<code class="filename"><em class="replaceable"><code>PREFIX</code></em>/etc</code> by default.
</p></dd><dt><span class="term"><code class="option">--libdir=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
Sets the location to install libraries and dynamically loadable
modules. The default is
<code class="filename"><em class="replaceable"><code>EXEC-PREFIX</code></em>/lib</code>.
</p></dd><dt><span class="term"><code class="option">--includedir=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
Sets the directory for installing C and C++ header files. The
default is <code class="filename"><em class="replaceable"><code>PREFIX</code></em>/include</code>.
</p></dd><dt><span class="term"><code class="option">--datarootdir=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
Sets the root directory for various types of read-only data
files. This only sets the default for some of the following
options. The default is
<code class="filename"><em class="replaceable"><code>PREFIX</code></em>/share</code>.
</p></dd><dt><span class="term"><code class="option">--datadir=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
Sets the directory for read-only data files used by the
installed programs. The default is
<code class="filename"><em class="replaceable"><code>DATAROOTDIR</code></em></code>. Note that this has
nothing to do with where your database files will be placed.
</p></dd><dt><span class="term"><code class="option">--localedir=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
Sets the directory for installing locale data, in particular
message translation catalog files. The default is
<code class="filename"><em class="replaceable"><code>DATAROOTDIR</code></em>/locale</code>.
</p></dd><dt><span class="term"><code class="option">--mandir=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
The man pages that come with <span class="productname">PostgreSQL</span> will be installed under
this directory, in their respective
<code class="filename">man<em class="replaceable"><code>x</code></em></code> subdirectories.
The default is <code class="filename"><em class="replaceable"><code>DATAROOTDIR</code></em>/man</code>.
</p></dd><dt><span class="term"><code class="option">--docdir=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
Sets the root directory for installing documentation files,
except <span class="quote">“<span class="quote">man</span>”</span> pages. This only sets the default for
the following options. The default value for this option is
<code class="filename"><em class="replaceable"><code>DATAROOTDIR</code></em>/doc/postgresql</code>.
</p></dd><dt><span class="term"><code class="option">--htmldir=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
The HTML-formatted documentation for
<span class="productname">PostgreSQL</span> will be installed under
this directory. The default is
<code class="filename"><em class="replaceable"><code>DATAROOTDIR</code></em></code>.
</p></dd></dl></div><p>
</p><div class="note"><h3 class="title">Note</h3><p>
Care has been taken to make it possible to install
<span class="productname">PostgreSQL</span> into shared installation locations
(such as <code class="filename">/usr/local/include</code>) without
interfering with the namespace of the rest of the system. First,
the string <span class="quote">“<span class="quote"><code class="literal">/postgresql</code></span>”</span> is
automatically appended to <code class="varname">datadir</code>,
<code class="varname">sysconfdir</code>, and <code class="varname">docdir</code>,
unless the fully expanded directory name already contains the
string <span class="quote">“<span class="quote"><code class="literal">postgres</code></span>”</span> or
<span class="quote">“<span class="quote"><code class="literal">pgsql</code></span>”</span>. For example, if you choose
<code class="filename">/usr/local</code> as prefix, the documentation will
be installed in <code class="filename">/usr/local/doc/postgresql</code>,
but if the prefix is <code class="filename">/opt/postgres</code>, then it
will be in <code class="filename">/opt/postgres/doc</code>. The public C
header files of the client interfaces are installed into
<code class="varname">includedir</code> and are namespace-clean. The
internal header files and the server header files are installed
into private directories under <code class="varname">includedir</code>. See
the documentation of each interface for information about how to
access its header files. Finally, a private subdirectory will
also be created, if appropriate, under <code class="varname">libdir</code>
for dynamically loadable modules.
</p></div><p>
</p><p>
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="option">--with-extra-version=<em class="replaceable"><code>STRING</code></em></code></span></dt><dd><p>
Append <em class="replaceable"><code>STRING</code></em> to the PostgreSQL version number. You
can use this, for example, to mark binaries built from unreleased Git
snapshots or containing custom patches with an extra version string
such as a <code class="command">git describe</code> identifier or a
distribution package release number.
</p></dd><dt><span class="term"><code class="option">--with-includes=<em class="replaceable"><code>DIRECTORIES</code></em></code></span></dt><dd><p>
<em class="replaceable"><code>DIRECTORIES</code></em> is a colon-separated list of
directories that will be added to the list the compiler
searches for header files. If you have optional packages
(such as GNU <span class="application">Readline</span>) installed in a non-standard
location,
you have to use this option and probably also the corresponding
<code class="option">--with-libraries</code> option.
</p><p>
Example: <code class="literal">--with-includes=/opt/gnu/include:/usr/sup/include</code>.
</p></dd><dt><span class="term"><code class="option">--with-libraries=<em class="replaceable"><code>DIRECTORIES</code></em></code></span></dt><dd><p>
<em class="replaceable"><code>DIRECTORIES</code></em> is a colon-separated list of
directories to search for libraries. You will probably have
to use this option (and the corresponding
<code class="option">--with-includes</code> option) if you have packages
installed in non-standard locations.
</p><p>
Example: <code class="literal">--with-libraries=/opt/gnu/lib:/usr/sup/lib</code>.
</p></dd><dt><span class="term"><code class="option">--enable-nls[<span class="optional">=<em class="replaceable"><code>LANGUAGES</code></em></span>]</code></span></dt><dd><p>
Enables Native Language Support (<acronym class="acronym">NLS</acronym>),
that is, the ability to display a program's messages in a
language other than English.
<em class="replaceable"><code>LANGUAGES</code></em> is an optional space-separated
list of codes of the languages that you want supported, for
example <code class="literal">--enable-nls='de fr'</code>. (The intersection
between your list and the set of actually provided
translations will be computed automatically.) If you do not
specify a list, then all available translations are
installed.
</p><p>
To use this option, you will need an implementation of the
<span class="application">Gettext</span> API; see above.
</p></dd><dt><span class="term"><code class="option">--with-pgport=<em class="replaceable"><code>NUMBER</code></em></code></span></dt><dd><p>
Set <em class="replaceable"><code>NUMBER</code></em> as the default port number for
server and clients. The default is 5432. The port can always
be changed later on, but if you specify it here then both
server and clients will have the same default compiled in,
which can be very convenient. Usually the only good reason
to select a non-default value is if you intend to run multiple
<span class="productname">PostgreSQL</span> servers on the same machine.
</p></dd><dt><span class="term"><code class="option">--with-perl</code></span></dt><dd><p>
Build the <span class="application">PL/Perl</span> server-side language.
</p></dd><dt><span class="term"><code class="option">--with-python</code></span></dt><dd><p>
Build the <span class="application">PL/Python</span> server-side language.
</p></dd><dt><span class="term"><code class="option">--with-tcl</code></span></dt><dd><p>
Build the <span class="application">PL/Tcl</span> server-side language.
</p></dd><dt><span class="term"><code class="option">--with-tclconfig=<em class="replaceable"><code>DIRECTORY</code></em></code></span></dt><dd><p>
Tcl installs the file <code class="filename">tclConfig.sh</code>, which
contains configuration information needed to build modules
interfacing to Tcl. This file is normally found automatically
at a well-known location, but if you want to use a different
version of Tcl you can specify the directory in which to look
for it.
</p></dd><dt><span class="term"><code class="option">--with-gssapi</code></span></dt><dd><p>
Build with support for GSSAPI authentication. On many
systems, the GSSAPI (usually a part of the Kerberos installation)
system is not installed in a location
that is searched by default (e.g., <code class="filename">/usr/include</code>,
<code class="filename">/usr/lib</code>), so you must use the options
<code class="option">--with-includes</code> and <code class="option">--with-libraries</code> in
addition to this option. <code class="filename">configure</code> will check
for the required header files and libraries to make sure that
your GSSAPI installation is sufficient before proceeding.
</p></dd><dt><span class="term"><code class="option">--with-krb-srvnam=<em class="replaceable"><code>NAME</code></em></code></span></dt><dd><p>
The default name of the Kerberos service principal used
by GSSAPI.
<code class="literal">postgres</code> is the default. There's usually no
reason to change this unless you have a Windows environment,
in which case it must be set to upper case
<code class="literal">POSTGRES</code>.
</p></dd><dt id="CONFIGURE-WITH-LLVM"><span class="term"><code class="option">--with-llvm</code></span></dt><dd><p>
Build with support for <span class="productname">LLVM</span> based
<acronym class="acronym">JIT</acronym> compilation<span class="phrase"> (see <a class="xref" href="jit.html" title="Chapter 32. Just-in-Time Compilation (JIT)">Chapter 32</a>)</span>. This
requires the <span class="productname">LLVM</span> library to be installed.
The minimum required version of <span class="productname">LLVM</span> is
currently 3.9.
</p><p>
<code class="command">llvm-config</code><a id="id-1.6.3.7.2.1.6.1.12.2.2.2" class="indexterm"></a>
will be used to find the required compilation options.
<code class="command">llvm-config</code>, and then
<code class="command">llvm-config-$major-$minor</code> for all supported
versions, will be searched on <code class="envar">PATH</code>. If that would not
yield the correct binary, use <code class="envar">LLVM_CONFIG</code> to specify a
path to the correct <code class="command">llvm-config</code>. For example
</p><pre class="programlisting">
./configure ... --with-llvm LLVM_CONFIG='/path/to/llvm/bin/llvm-config'
</pre><p>
</p><p>
<span class="productname">LLVM</span> support requires a compatible
<code class="command">clang</code> compiler (specified, if necessary, using the
<code class="envar">CLANG</code> environment variable), and a working C++
compiler (specified, if necessary, using the <code class="envar">CXX</code>
environment variable).
</p></dd><dt><span class="term"><code class="option">--with-icu</code></span></dt><dd><p>
Build with support for
the <span class="productname">ICU</span><a id="id-1.6.3.7.2.1.6.1.13.2.1.2" class="indexterm"></a>
library. This requires the <span class="productname">ICU4C</span> package
to be installed. The minimum required version
of <span class="productname">ICU4C</span> is currently 4.2.
</p><p>
By default,
<span class="productname">pkg-config</span><a id="id-1.6.3.7.2.1.6.1.13.2.2.2" class="indexterm"></a>
will be used to find the required compilation options. This is
supported for <span class="productname">ICU4C</span> version 4.6 and later.
For older versions, or if <span class="productname">pkg-config</span> is
not available, the variables <code class="envar">ICU_CFLAGS</code>
and <code class="envar">ICU_LIBS</code> can be specified
to <code class="filename">configure</code>, like in this example:
</p><pre class="programlisting">
./configure ... --with-icu ICU_CFLAGS='-I/some/where/include' ICU_LIBS='-L/some/where/lib -licui18n -licuuc -licudata'
</pre><p>
(If <span class="productname">ICU4C</span> is in the default search path
for the compiler, then you still need to specify a nonempty string in
order to avoid use of <span class="productname">pkg-config</span>, for
example, <code class="literal">ICU_CFLAGS=' '</code>.)
</p></dd><dt><span class="term"><code class="option">--with-openssl</code>
<a id="id-1.6.3.7.2.1.6.1.14.1.2" class="indexterm"></a>
</span></dt><dd><p>
Build with support for <acronym class="acronym">SSL</acronym> (encrypted)
connections. This requires the <span class="productname">OpenSSL</span>
package to be installed. <code class="filename">configure</code> will check
for the required header files and libraries to make sure that
your <span class="productname">OpenSSL</span> installation is sufficient
before proceeding.
</p></dd><dt><span class="term"><code class="option">--with-pam</code></span></dt><dd><p>
Build with <acronym class="acronym">PAM</acronym><a id="id-1.6.3.7.2.1.6.1.15.2.1.2" class="indexterm"></a>
(Pluggable Authentication Modules) support.
</p></dd><dt><span class="term"><code class="option">--with-bsd-auth</code></span></dt><dd><p>
Build with BSD Authentication support.
(The BSD Authentication framework is
currently only available on OpenBSD.)
</p></dd><dt><span class="term"><code class="option">--with-ldap</code></span></dt><dd><p>
Build with <acronym class="acronym">LDAP</acronym><a id="id-1.6.3.7.2.1.6.1.17.2.1.2" class="indexterm"></a>
support for authentication and connection parameter lookup (see
<span id="INSTALL-LDAP-LINKS" class="phrase"><a class="xref" href="libpq-ldap.html" title="34.17. LDAP Lookup of Connection Parameters">Section 34.17</a> and
<a class="xref" href="auth-ldap.html" title="20.10. LDAP Authentication">Section 20.10</a></span> for more information). On Unix,
this requires the <span class="productname">OpenLDAP</span> package to be
installed. On Windows, the default <span class="productname">WinLDAP</span>
library is used. <code class="filename">configure</code> will check for the required
header files and libraries to make sure that your
<span class="productname">OpenLDAP</span> installation is sufficient before
proceeding.
</p></dd><dt><span class="term"><code class="option">--with-systemd</code></span></dt><dd><p>
Build with support
for <span class="application">systemd</span><a id="id-1.6.3.7.2.1.6.1.18.2.1.2" class="indexterm"></a>
service notifications. This improves integration if the server binary
is started under <span class="application">systemd</span> but has no impact
otherwise<span class="phrase">; see <a class="xref" href="server-start.html" title="18.3. Starting the Database Server">Section 18.3</a> for more
information</span>. <span class="application">libsystemd</span> and the
associated header files need to be installed to be able to use this
option.
</p></dd><dt><span class="term"><code class="option">--without-readline</code></span></dt><dd><p>
Prevents use of the <span class="application">Readline</span> library
(and <span class="application">libedit</span> as well). This option disables
command-line editing and history in
<span class="application">psql</span>, so it is not recommended.
</p></dd><dt><span class="term"><code class="option">--with-libedit-preferred</code></span></dt><dd><p>
Favors the use of the BSD-licensed <span class="application">libedit</span> library
rather than GPL-licensed <span class="application">Readline</span>. This option
is significant only if you have both libraries installed; the
default in that case is to use <span class="application">Readline</span>.
</p></dd><dt><span class="term"><code class="option">--with-bonjour</code></span></dt><dd><p>
Build with Bonjour support. This requires Bonjour support
in your operating system. Recommended on macOS.
</p></dd><dt><span class="term"><code class="option">--with-uuid=<em class="replaceable"><code>LIBRARY</code></em></code></span></dt><dd><p>
Build the <a class="xref" href="uuid-ossp.html" title="F.44. uuid-ossp">uuid-ossp</a> module
(which provides functions to generate UUIDs), using the specified
UUID library.<a id="id-1.6.3.7.2.1.6.1.22.2.1.2" class="indexterm"></a>
<em class="replaceable"><code>LIBRARY</code></em> must be one of:
</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p>
<code class="option">bsd</code> to use the UUID functions found in FreeBSD, NetBSD,
and some other BSD-derived systems
</p></li><li class="listitem"><p>
<code class="option">e2fs</code> to use the UUID library created by
the <code class="literal">e2fsprogs</code> project; this library is present in most
Linux systems and in macOS, and can be obtained for other
platforms as well
</p></li><li class="listitem"><p>
<code class="option">ossp</code> to use the <a class="ulink" href="http://www.ossp.org/pkg/lib/uuid/" target="_top">OSSP UUID library</a>
</p></li></ul></div></dd><dt><span class="term"><code class="option">--with-ossp-uuid</code></span></dt><dd><p>
Obsolete equivalent of <code class="literal">--with-uuid=ossp</code>.
</p></dd><dt><span class="term"><code class="option">--with-libxml</code></span></dt><dd><p>
Build with libxml (enables SQL/XML support). Libxml version 2.6.23 or
later is required for this feature.
</p><p>
Libxml installs a program <code class="command">xml2-config</code> that
can be used to detect the required compiler and linker
options. PostgreSQL will use it automatically if found. To
specify a libxml installation at an unusual location, you can
either set the environment variable
<code class="envar">XML2_CONFIG</code> to point to the
<code class="command">xml2-config</code> program belonging to the
installation, or use the options
<code class="option">--with-includes</code> and
<code class="option">--with-libraries</code>.
</p></dd><dt><span class="term"><code class="option">--with-libxslt</code></span></dt><dd><p>
Use libxslt when building the
<a class="xref" href="xml2.html" title="F.45. xml2">xml2</a>
module. <span class="application">xml2</span> relies on this library
to perform XSL transformations of XML.
</p></dd><dt><span class="term"><code class="option">--disable-float4-byval</code></span></dt><dd><p>
Disable passing float4 values <span class="quote">“<span class="quote">by value</span>”</span>, causing them
to be passed <span class="quote">“<span class="quote">by reference</span>”</span> instead. This option costs
performance, but may be needed for compatibility with old
user-defined functions that are written in C and use the
<span class="quote">“<span class="quote">version 0</span>”</span> calling convention. A better long-term
solution is to update any such functions to use the
<span class="quote">“<span class="quote">version 1</span>”</span> calling convention.
</p></dd><dt><span class="term"><code class="option">--disable-float8-byval</code></span></dt><dd><p>
Disable passing float8 values <span class="quote">“<span class="quote">by value</span>”</span>, causing them
to be passed <span class="quote">“<span class="quote">by reference</span>”</span> instead. This option costs
performance, but may be needed for compatibility with old
user-defined functions that are written in C and use the
<span class="quote">“<span class="quote">version 0</span>”</span> calling convention. A better long-term
solution is to update any such functions to use the
<span class="quote">“<span class="quote">version 1</span>”</span> calling convention.
Note that this option affects not only float8, but also int8 and some
related types such as timestamp.
On 32-bit platforms, <code class="option">--disable-float8-byval</code> is the default
and it is not allowed to select <code class="option">--enable-float8-byval</code>.
</p></dd><dt><span class="term"><code class="option">--with-segsize=<em class="replaceable"><code>SEGSIZE</code></em></code></span></dt><dd><p>
Set the <em class="firstterm">segment size</em>, in gigabytes. Large tables are
divided into multiple operating-system files, each of size equal
to the segment size. This avoids problems with file size limits
that exist on many platforms. The default segment size, 1 gigabyte,
is safe on all supported platforms. If your operating system has
<span class="quote">“<span class="quote">largefile</span>”</span> support (which most do, nowadays), you can use
a larger segment size. This can be helpful to reduce the number of
file descriptors consumed when working with very large tables.
But be careful not to select a value larger than is supported
by your platform and the file systems you intend to use. Other
tools you might wish to use, such as <span class="application">tar</span>, could
also set limits on the usable file size.
It is recommended, though not absolutely required, that this value
be a power of 2.
Note that changing this value requires an initdb.
</p></dd><dt><span class="term"><code class="option">--with-blocksize=<em class="replaceable"><code>BLOCKSIZE</code></em></code></span></dt><dd><p>
Set the <em class="firstterm">block size</em>, in kilobytes. This is the unit
of storage and I/O within tables. The default, 8 kilobytes,
is suitable for most situations; but other values may be useful
in special cases.
The value must be a power of 2 between 1 and 32 (kilobytes).
Note that changing this value requires an initdb.
</p></dd><dt><span class="term"><code class="option">--with-wal-blocksize=<em class="replaceable"><code>BLOCKSIZE</code></em></code></span></dt><dd><p>
Set the <em class="firstterm">WAL block size</em>, in kilobytes. This is the unit
of storage and I/O within the WAL log. The default, 8 kilobytes,
is suitable for most situations; but other values may be useful
in special cases.
The value must be a power of 2 between 1 and 64 (kilobytes).
Note that changing this value requires an initdb.
</p></dd><dt><span class="term"><code class="option">--disable-spinlocks</code></span></dt><dd><p>
Allow the build to succeed even if <span class="productname">PostgreSQL</span>
has no CPU spinlock support for the platform. The lack of
spinlock support will result in poor performance; therefore,
this option should only be used if the build aborts and
informs you that the platform lacks spinlock support. If this
option is required to build <span class="productname">PostgreSQL</span> on
your platform, please report the problem to the
<span class="productname">PostgreSQL</span> developers.
</p></dd><dt><span class="term"><code class="option">--disable-strong-random</code></span></dt><dd><p>
Allow the build to succeed even if <span class="productname">PostgreSQL</span>
has no support for strong random numbers on the platform.
A source of random numbers is needed for some authentication
protocols, as well as some routines in the
<a class="xref" href="pgcrypto.html" title="F.25. pgcrypto">pgcrypto</a>
module. <code class="option">--disable-strong-random</code> disables functionality that
requires cryptographically strong random numbers, and substitutes
a weak pseudo-random-number-generator for the generation of
authentication salt values and query cancel keys. It may make
authentication less secure.
</p></dd><dt><span class="term"><code class="option">--disable-thread-safety</code></span></dt><dd><p>
Disable the thread-safety of client libraries. This prevents
concurrent threads in <span class="application">libpq</span> and
<span class="application">ECPG</span> programs from safely controlling
their private connection handles.
</p></dd><dt><span class="term"><code class="option">--with-system-tzdata=<em class="replaceable"><code>DIRECTORY</code></em></code>
<a id="id-1.6.3.7.2.1.6.1.34.1.2" class="indexterm"></a>
</span></dt><dd><p>
<span class="productname">PostgreSQL</span> includes its own time zone database,
which it requires for date and time operations. This time zone
database is in fact compatible with the IANA time zone
database provided by many operating systems such as FreeBSD,
Linux, and Solaris, so it would be redundant to install it again.
When this option is used, the system-supplied time zone database
in <em class="replaceable"><code>DIRECTORY</code></em> is used instead of the one
included in the PostgreSQL source distribution.
<em class="replaceable"><code>DIRECTORY</code></em> must be specified as an
absolute path. <code class="filename">/usr/share/zoneinfo</code> is a
likely directory on some operating systems. Note that the
installation routine will not detect mismatching or erroneous time
zone data. If you use this option, you are advised to run the
regression tests to verify that the time zone data you have
pointed to works correctly with <span class="productname">PostgreSQL</span>.
</p><a id="id-1.6.3.7.2.1.6.1.34.2.2" class="indexterm"></a><p>
This option is mainly aimed at binary package distributors
who know their target operating system well. The main
advantage of using this option is that the PostgreSQL package
won't need to be upgraded whenever any of the many local
daylight-saving time rules change. Another advantage is that
PostgreSQL can be cross-compiled more straightforwardly if the
time zone database files do not need to be built during the
installation.
</p></dd><dt><span class="term"><code class="option">--without-zlib</code></span></dt><dd><p>
<a id="id-1.6.3.7.2.1.6.1.35.2.1.1" class="indexterm"></a>
Prevents use of the <span class="application">Zlib</span> library. This disables
support for compressed archives in <span class="application">pg_dump</span>
and <span class="application">pg_restore</span>.
This option is only intended for those rare systems where this
library is not available.
</p></dd><dt><span class="term"><code class="option">--enable-debug</code></span></dt><dd><p>
Compiles all programs and libraries with debugging symbols.
This means that you can run the programs in a debugger
to analyze problems. This enlarges the size of the installed
executables considerably, and on non-GCC compilers it usually
also disables compiler optimization, causing slowdowns. However,
having the symbols available is extremely helpful for dealing
with any problems that might arise. Currently, this option is
recommended for production installations only if you use GCC.
But you should always have it on if you are doing development work
or running a beta version.
</p></dd><dt><span class="term"><code class="option">--enable-coverage</code></span></dt><dd><p>
If using GCC, all programs and libraries are compiled with
code coverage testing instrumentation. When run, they
generate files in the build directory with code coverage
metrics.
<span class="phrase">See <a class="xref" href="regress-coverage.html" title="33.5. Test Coverage Examination">Section 33.5</a>
for more information.</span> This option is for use only with GCC
and when doing development work.
</p></dd><dt><span class="term"><code class="option">--enable-profiling</code></span></dt><dd><p>
If using GCC, all programs and libraries are compiled so they
can be profiled. On backend exit, a subdirectory will be created
that contains the <code class="filename">gmon.out</code> file for use in profiling.
This option is for use only with GCC and when doing development work.
</p></dd><dt><span class="term"><code class="option">--enable-cassert</code></span></dt><dd><p>
Enables <em class="firstterm">assertion</em> checks in the server, which test for
many <span class="quote">“<span class="quote">cannot happen</span>”</span> conditions. This is invaluable for
code development purposes, but the tests can slow down the
server significantly.
Also, having the tests turned on won't necessarily enhance the
stability of your server! The assertion checks are not categorized
for severity, and so what might be a relatively harmless bug will
still lead to server restarts if it triggers an assertion
failure. This option is not recommended for production use, but
you should have it on for development work or when running a beta
version.
</p></dd><dt><span class="term"><code class="option">--enable-depend</code></span></dt><dd><p>
Enables automatic dependency tracking. With this option, the
makefiles are set up so that all affected object files will
be rebuilt when any header file is changed. This is useful
if you are doing development work, but is just wasted overhead
if you intend only to compile once and install. At present,
this option only works with GCC.
</p></dd><dt><span class="term"><code class="option">--enable-dtrace</code></span></dt><dd><p>
<a id="id-1.6.3.7.2.1.6.1.41.2.1.1" class="indexterm"></a>
Compiles <span class="productname">PostgreSQL</span> with support for the
dynamic tracing tool DTrace.
<span class="phrase">See <a class="xref" href="dynamic-trace.html" title="28.5. Dynamic Tracing">Section 28.5</a>
for more information.</span>
</p><p>
To point to the <code class="command">dtrace</code> program, the
environment variable <code class="envar">DTRACE</code> can be set. This
will often be necessary because <code class="command">dtrace</code> is
typically installed under <code class="filename">/usr/sbin</code>,
which might not be in the path.
</p><p>
Extra command-line options for the <code class="command">dtrace</code> program
can be specified in the environment variable
<code class="envar">DTRACEFLAGS</code>. On Solaris,
to include DTrace support in a 64-bit binary, you must specify
<code class="literal">DTRACEFLAGS="-64"</code> to configure. For example,
using the GCC compiler:
</p><pre class="screen">
./configure CC='gcc -m64' --enable-dtrace DTRACEFLAGS='-64' ...
</pre><p>
Using Sun's compiler:
</p><pre class="screen">
./configure CC='/opt/SUNWspro/bin/cc -xtarget=native64' --enable-dtrace DTRACEFLAGS='-64' ...
</pre><p>
</p></dd><dt><span class="term"><code class="option">--enable-tap-tests</code></span></dt><dd><p>
Enable tests using the Perl TAP tools. This requires a Perl
installation and the Perl module <code class="literal">IPC::Run</code>.
<span class="phrase">See <a class="xref" href="regress-tap.html" title="33.4. TAP Tests">Section 33.4</a> for more information.</span>
</p></dd></dl></div><p>
</p><p>
If you prefer a C compiler different from the one
<code class="filename">configure</code> picks, you can set the
environment variable <code class="envar">CC</code> to the program of your choice.
By default, <code class="filename">configure</code> will pick
<code class="filename">gcc</code> if available, else the platform's
default (usually <code class="filename">cc</code>). Similarly, you can override the
default compiler flags if needed with the <code class="envar">CFLAGS</code> variable.
</p><p>
You can specify environment variables on the
<code class="filename">configure</code> command line, for example:
</p><pre class="screen">
<strong class="userinput"><code>./configure CC=/opt/bin/gcc CFLAGS='-O2 -pipe'</code></strong>
</pre><p>
</p><p>
Here is a list of the significant variables that can be set in
this manner:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="envar">BISON</code></span></dt><dd><p>
Bison program
</p></dd><dt><span class="term"><code class="envar">CC</code></span></dt><dd><p>
C compiler
</p></dd><dt><span class="term"><code class="envar">CFLAGS</code></span></dt><dd><p>
options to pass to the C compiler
</p></dd><dt><span class="term"><code class="envar">CLANG</code></span></dt><dd><p>
path to <code class="command">clang</code> program used to process source code
for inlining when compiling with <code class="literal">--with-llvm</code>
</p></dd><dt><span class="term"><code class="envar">CPP</code></span></dt><dd><p>
C preprocessor
</p></dd><dt><span class="term"><code class="envar">CPPFLAGS</code></span></dt><dd><p>
options to pass to the C preprocessor
</p></dd><dt><span class="term"><code class="envar">CXX</code></span></dt><dd><p>
C++ compiler
</p></dd><dt><span class="term"><code class="envar">CXXFLAGS</code></span></dt><dd><p>
options to pass to the C++ compiler
</p></dd><dt><span class="term"><code class="envar">DTRACE</code></span></dt><dd><p>
location of the <code class="command">dtrace</code> program
</p></dd><dt><span class="term"><code class="envar">DTRACEFLAGS</code></span></dt><dd><p>
options to pass to the <code class="command">dtrace</code> program
</p></dd><dt><span class="term"><code class="envar">FLEX</code></span></dt><dd><p>
Flex program
</p></dd><dt><span class="term"><code class="envar">LDFLAGS</code></span></dt><dd><p>
options to use when linking either executables or shared libraries
</p></dd><dt><span class="term"><code class="envar">LDFLAGS_EX</code></span></dt><dd><p>
additional options for linking executables only
</p></dd><dt><span class="term"><code class="envar">LDFLAGS_SL</code></span></dt><dd><p>
additional options for linking shared libraries only
</p></dd><dt><span class="term"><code class="envar">LLVM_CONFIG</code></span></dt><dd><p>
<code class="command">llvm-config</code> program used to locate the
<span class="productname">LLVM</span> installation.
</p></dd><dt><span class="term"><code class="envar">MSGFMT</code></span></dt><dd><p>
<code class="command">msgfmt</code> program for native language support
</p></dd><dt><span class="term"><code class="envar">PERL</code></span></dt><dd><p>
Perl interpreter program. This will be used to determine the
dependencies for building PL/Perl. The default is
<code class="command">perl</code>.
</p></dd><dt><span class="term"><code class="envar">PYTHON</code></span></dt><dd><p>
Python interpreter program. This will be used to
determine the dependencies for building PL/Python. Also,
whether Python 2 or 3 is specified here (or otherwise
implicitly chosen) determines which variant of the PL/Python
language becomes available. See
<a class="xref" href="plpython-python23.html" title="46.1. Python 2 vs. Python 3">Section 46.1</a>
for more information. If this is not set, the following are probed
in this order: <code class="literal">python python3 python2</code>.
</p></dd><dt><span class="term"><code class="envar">TCLSH</code></span></dt><dd><p>
Tcl interpreter program. This will be used to
determine the dependencies for building PL/Tcl, and it will
be substituted into Tcl scripts.
</p></dd><dt><span class="term"><code class="envar">XML2_CONFIG</code></span></dt><dd><p>
<code class="command">xml2-config</code> program used to locate the
libxml installation.
</p></dd></dl></div><p>
</p><p>
Sometimes it is useful to add compiler flags after-the-fact to the set
that were chosen by <code class="filename">configure</code>. An important example is
that <span class="application">gcc</span>'s <code class="option">-Werror</code> option cannot be included
in the <code class="envar">CFLAGS</code> passed to <code class="filename">configure</code>, because
it will break many of <code class="filename">configure</code>'s built-in tests. To add
such flags, include them in the <code class="envar">COPT</code> environment variable
while running <code class="filename">make</code>. The contents of <code class="envar">COPT</code>
are added to both the <code class="envar">CFLAGS</code> and <code class="envar">LDFLAGS</code>
options set up by <code class="filename">configure</code>. For example, you could do
</p><pre class="screen">
<strong class="userinput"><code>make COPT='-Werror'</code></strong>
</pre><p>
or
</p><pre class="screen">
<strong class="userinput"><code>export COPT='-Werror'</code></strong>
<strong class="userinput"><code>make</code></strong>
</pre><p>
</p><div class="note"><h3 class="title">Note</h3><p>
When developing code inside the server, it is recommended to
use the configure options <code class="option">--enable-cassert</code> (which
turns on many run-time error checks) and <code class="option">--enable-debug</code>
(which improves the usefulness of debugging tools).
</p><p>
If using GCC, it is best to build with an optimization level of
at least <code class="option">-O1</code>, because using no optimization
(<code class="option">-O0</code>) disables some important compiler warnings (such
as the use of uninitialized variables). However, non-zero
optimization levels can complicate debugging because stepping
through compiled code will usually not match up one-to-one with
source code lines. If you get confused while trying to debug
optimized code, recompile the specific files of interest with
<code class="option">-O0</code>. An easy way to do this is by passing an option
to <span class="application">make</span>: <code class="command">make PROFILE=-O0 file.o</code>.
</p><p>
The <code class="envar">COPT</code> and <code class="envar">PROFILE</code> environment variables are
actually handled identically by the <span class="productname">PostgreSQL</span>
makefiles. Which to use is a matter of preference, but a common habit
among developers is to use <code class="envar">PROFILE</code> for one-time flag
adjustments, while <code class="envar">COPT</code> might be kept set all the time.
</p></div></li><li class="step" id="BUILD"><p class="title"><strong>Build</strong></p><p>
To start the build, type either of:
</p><pre class="screen">
<strong class="userinput"><code>make</code></strong>
<strong class="userinput"><code>make all</code></strong>
</pre><p>
(Remember to use <acronym class="acronym">GNU</acronym> <span class="application">make</span>.)
The build will take a few minutes depending on your
hardware. The last line displayed should be:
</p><pre class="screen">
All of PostgreSQL successfully made. Ready to install.
</pre><p>
</p><p>
If you want to build everything that can be built, including the
documentation (HTML and man pages), and the additional modules
(<code class="filename">contrib</code>), type instead:
</p><pre class="screen">
<strong class="userinput"><code>make world</code></strong>
</pre><p>
The last line displayed should be:
</p><pre class="screen">
PostgreSQL, contrib, and documentation successfully made. Ready to install.
</pre><p>
</p><p>
If you want to invoke the build from another makefile rather than
manually, you must unset <code class="varname">MAKELEVEL</code> or set it to zero,
for instance like this:
</p><pre class="programlisting">
build-postgresql:
$(MAKE) -C postgresql MAKELEVEL=0 all
</pre><p>
Failure to do that can lead to strange error messages, typically about
missing header files.
</p></li><li class="step"><p class="title"><strong>Regression Tests</strong></p><a id="id-1.6.3.7.2.3.2" class="indexterm"></a><p>
If you want to test the newly built server before you install it,
you can run the regression tests at this point. The regression
tests are a test suite to verify that <span class="productname">PostgreSQL</span>
runs on your machine in the way the developers expected it
to. Type:
</p><pre class="screen">
<strong class="userinput"><code>make check</code></strong>
</pre><p>
(This won't work as root; do it as an unprivileged user.)
See <a class="xref" href="regress.html" title="Chapter 33. Regression Tests">Chapter 33</a> for
detailed information about interpreting the test results. You can
repeat this test at any later time by issuing the same command.
</p></li><li class="step" id="INSTALL"><p class="title"><strong>Installing the Files</strong></p><div class="note"><h3 class="title">Note</h3><p>
If you are upgrading an existing system be sure to read
<a class="xref" href="upgrading.html" title="18.6. Upgrading a PostgreSQL Cluster">Section 18.6</a>,
which has instructions about upgrading a
cluster.
</p></div><p>
To install <span class="productname">PostgreSQL</span> enter:
</p><pre class="screen">
<strong class="userinput"><code>make install</code></strong>
</pre><p>
This will install files into the directories that were specified
in <a class="xref" href="install-procedure.html#CONFIGURE" title="Configuration">Step 1</a>. Make sure that you have appropriate
permissions to write into that area. Normally you need to do this
step as root. Alternatively, you can create the target
directories in advance and arrange for appropriate permissions to
be granted.
</p><p>
To install the documentation (HTML and man pages), enter:
</p><pre class="screen">
<strong class="userinput"><code>make install-docs</code></strong>
</pre><p>
</p><p>
If you built the world above, type instead:
</p><pre class="screen">
<strong class="userinput"><code>make install-world</code></strong>
</pre><p>
This also installs the documentation.
</p><p>
You can use <code class="literal">make install-strip</code> instead of
<code class="literal">make install</code> to strip the executable files and
libraries as they are installed. This will save some space. If
you built with debugging support, stripping will effectively
remove the debugging support, so it should only be done if
debugging is no longer needed. <code class="literal">install-strip</code>
tries to do a reasonable job saving space, but it does not have
perfect knowledge of how to strip every unneeded byte from an
executable file, so if you want to save all the disk space you
possibly can, you will have to do manual work.
</p><p>
The standard installation provides all the header files needed for client
application development as well as for server-side program
development, such as custom functions or data types written in C.
(Prior to <span class="productname">PostgreSQL</span> 8.0, a separate <code class="literal">make
install-all-headers</code> command was needed for the latter, but this
step has been folded into the standard install.)
</p><p><strong>Client-only installation: </strong>
If you want to install only the client applications and
interface libraries, then you can use these commands:
</p><pre class="screen">
<strong class="userinput"><code>make -C src/bin install</code></strong>
<strong class="userinput"><code>make -C src/include install</code></strong>
<strong class="userinput"><code>make -C src/interfaces install</code></strong>
<strong class="userinput"><code>make -C doc install</code></strong>
</pre><p>
<code class="filename">src/bin</code> has a few binaries for server-only use,
but they are small.
</p></li></ol></div><p><strong>Uninstallation: </strong>
To undo the installation use the command <code class="command">make
uninstall</code>. However, this will not remove any created directories.
</p><p><strong>Cleaning: </strong>
After the installation you can free disk space by removing the built
files from the source tree with the command <code class="command">make
clean</code>. This will preserve the files made by the <code class="command">configure</code>
program, so that you can rebuild everything with <code class="command">make</code>
later on. To reset the source tree to the state in which it was
distributed, use <code class="command">make distclean</code>. If you are going to
build for several platforms within the same source tree you must do
this and re-configure for each platform. (Alternatively, use
a separate build tree for each platform, so that the source tree
remains unmodified.)
</p><p>
If you perform a build and then discover that your <code class="command">configure</code>
options were wrong, or if you change anything that <code class="command">configure</code>
investigates (for example, software upgrades), then it's a good
idea to do <code class="command">make distclean</code> before reconfiguring and
rebuilding. Without this, your changes in configuration choices
might not propagate everywhere they need to.
</p></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="install-getsource.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="installation.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="install-post.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">16.3. Getting The Source </td><td width="20%" align="center"><a accesskey="h" href="index.html">Home</a></td><td width="40%" align="right" valign="top"> 16.5. Post-Installation Setup</td></tr></table></div></body></html>
