PostgreSQL源码安装文档

This document describes the installation of PostgreSQL using the source
    code distribution. (If you are installing a pre-packaged distribution,
    such as an RPM or Debian package, ignore this document and read the
    packager's instructions instead.)
      __________________________________________________________________

Short Version

./configure

make

su

make install

adduser postgres

mkdir /usr/local/pgsql/data

chown postgres /usr/local/pgsql/data

su - postgres

/usr/local/pgsql/bin/initdb -D /usr/local/pgsql/data

/usr/local/pgsql/bin/postgres -D /usr/local/pgsql/data >logfile 2>&1 &

/usr/local/pgsql/bin/createdb test

/usr/local/pgsql/bin/psql test

The long version is the rest of this document.
      __________________________________________________________________

Requirements

In general, a modern Unix-compatible platform should be able to run
    PostgreSQL. The platforms that had received specific testing at the
    time of release are listed in the Section called Supported Platforms
    below. In the "doc" subdirectory of the distribution there are several
    platform-specific FAQ documents you might wish to consult if you are
    having trouble.

The following software packages are required for building PostgreSQL:

* GNU make version 3.80 or newer is required; other make programs or
        older GNU make versions will *not* work. (GNU make is sometimes
        installed under the name "gmake".) To test for GNU make enter:

make --version
      * You need an ISO/ANSI C compiler (at least C89-compliant). Recent
        versions of GCC are recommended, but PostgreSQL is known to build
        using a wide variety of compilers from different vendors.
      * tar is required to unpack the source distribution, in addition to
        either gzip or bzip2.
      * The GNU Readline library is used by default. It allows psql (the
        PostgreSQL command line SQL interpreter) to remember each command
        you type, and allows you to use arrow keys to recall and edit
        previous commands. This is very helpful and is strongly
        recommended. If you don't want to use it then you must specify the
        "--without-readline" option to "configure". As an alternative, you
        can often use the BSD-licensed "libedit" library, originally
        developed on NetBSD. The "libedit" library is GNU
        Readline-compatible and is used if "libreadline" is not found, or
        if "--with-libedit-preferred" is used as an option to "configure".
        If you are using a package-based Linux distribution, be aware that
        you need both the readline and readline-devel packages, if those
        are separate in your distribution.
      * The zlib compression library is used by default. If you don't want
        to use it then you must specify the "--without-zlib" option to
        "configure". Using this option disables support for compressed
        archives in pg_dump and pg_restore.

The following packages are optional. They are not required in the
    default configuration, but they are needed when certain build options
    are enabled, as explained below:

* To build the server programming language PL/Perl you need a full
        Perl installation, including the "libperl" library and the header
        files. Since PL/Perl will be a shared library, the "libperl"
        library must be a shared library also on most platforms. This
        appears to be the default in recent Perl versions, but it was not
        in earlier versions, and in any case it is the choice of whomever
        installed Perl at your site. "configure" will fail if building
        PL/Perl is selected but it cannot find a shared "libperl". In that
        case, you will have to rebuild and install Perl manually to be able
        to build PL/Perl. During the configuration process for Perl,
        request a shared library.
        If you intend to make more than incidental use of PL/Perl, you
        should ensure that the Perl installation was built with the
        usemultiplicity option enabled (perl -V will show whether this is
        the case).
      * To build the PL/Python server programming language, you need a
        Python installation with the header files and the distutils module.
        The minimum required version is Python 2.3. (To work with function
        arguments of type numeric, a 2.3.x installation must include the
        separately-available "cdecimal" module; note the PL/Python
        regression tests will not pass if that is missing.) Python 3 is
        supported if it's version 3.1 or later; but see the PL/Python
        documentation when using Python 3.
        Since PL/Python will be a shared library, the "libpython" library
        must be a shared library also on most platforms. This is not the
        case in a default Python installation built from source, but a
        shared library is available in many operating system distributions.
        "configure" will fail if building PL/Python is selected but it
        cannot find a shared "libpython". That might mean that you either
        have to install additional packages or rebuild (part of) your
        Python installation to provide this shared library. When building
        from source, run Python's configure with the --enable-shared flag.
      * To build the PL/Tcl procedural language, you of course need a Tcl
        installation. If you are using a pre-8.4 release of Tcl, ensure
        that it was built without multithreading support.
      * To enable Native Language Support (NLS), that is, the ability to
        display a program's messages in a language other than English, you
        need an implementation of the Gettext API. Some operating systems
        have this built-in (e.g., Linux, NetBSD, Solaris), for other
        systems you can download an add-on package from
        http://www.gnu.org/software/gettext/. If you are using the Gettext
        implementation in the GNU C library then you will additionally need
        the GNU Gettext package for some utility programs. For any of the
        other implementations you will not need it.
      * You need Kerberos, OpenSSL, OpenLDAP, and/or PAM, if you want to
        support authentication or encryption using those services.
      * To build the PostgreSQL documentation, there is a separate set of
        requirements; see the main documentation's appendix on
        documentation.

If you are building from a Git tree instead of using a released source
    package, or if you want to do server development, you also need the
    following packages:

* GNU Flex and Bison are needed to build from a Git checkout, or if
        you changed the actual scanner and parser definition files. If you
        need them, be sure to get Flex 2.5.31 or later and Bison 1.875 or
        later. Other lex and yacc programs cannot be used.
      * Perl 5.8 or later is needed to build from a Git checkout, or if you
        changed the input files for any of the build steps that use Perl
        scripts. If building on Windows you will need Perl in any case.
        Perl is also required to run some test suites.

If you need to get a GNU package, you can find it at your local GNU
    mirror site (see http://www.gnu.org/order/ftp.html for a list) or at
    ftp://ftp.gnu.org/gnu/.

Also check that you have sufficient disk space. You will need about 100
    MB for the source tree during compilation and about 20 MB for the
    installation directory. An empty database cluster takes about 35 MB;
    databases take about five times the amount of space that a flat text
    file with the same data would take. If you are going to run the
    regression tests you will temporarily need up to an extra 150 MB. Use
    the "df" command to check free disk space.
      __________________________________________________________________

Installation Procedure

1. Configuration
        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 "configure" script. For a default
        installation simply enter:

./configure
        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 "configure" in
        a directory outside the source tree, if you want to keep the build
        directory separate. This procedure is also called a VPATH build.
        Here's how:

mkdir build_dir

cd build_dir

/path/to/source/tree/configure [options go here]

make
        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 "/usr/local/pgsql" by
        default.
        You can customize the build and installation process by supplying
        one or more of the following command line options to "configure":

--prefix=PREFIX
                 Install all files under the directory "PREFIX" instead of
                 "/usr/local/pgsql". The actual files will be installed
                 into various subdirectories; no files will ever be
                 installed directly into the "PREFIX" directory.

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 man and doc locations
                 are not affected by this.)

For relocatable installs, you might want to use
                 "configure"'s --disable-rpath option. Also, you will need
                 to tell the operating system how to find the shared
                 libraries.

--exec-prefix=EXEC-PREFIX
                 You can install architecture-dependent files under a
                 different prefix, "EXEC-PREFIX", than what "PREFIX" was
                 set to. This can be useful to share
                 architecture-independent files between hosts. If you omit
                 this, then "EXEC-PREFIX" is set equal to "PREFIX" and both
                 architecture-dependent and independent files will be
                 installed under the same tree, which is probably what you
                 want.

--bindir=DIRECTORY
                 Specifies the directory for executable programs. The
                 default is "EXEC-PREFIX/bin", which normally means
                 "/usr/local/pgsql/bin".

--sysconfdir=DIRECTORY
                 Sets the directory for various configuration files,
                 "PREFIX/etc" by default.

--libdir=DIRECTORY
                 Sets the location to install libraries and dynamically
                 loadable modules. The default is "EXEC-PREFIX/lib".

--includedir=DIRECTORY
                 Sets the directory for installing C and C++ header files.
                 The default is "PREFIX/include".

--datarootdir=DIRECTORY
                 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 "PREFIX/share".

--datadir=DIRECTORY
                 Sets the directory for read-only data files used by the
                 installed programs. The default is "DATAROOTDIR". Note
                 that this has nothing to do with where your database files
                 will be placed.

--localedir=DIRECTORY
                 Sets the directory for installing locale data, in
                 particular message translation catalog files. The default
                 is "DATAROOTDIR/locale".

--mandir=DIRECTORY
                 The man pages that come with PostgreSQL will be installed
                 under this directory, in their respective "manx"
                 subdirectories. The default is "DATAROOTDIR/man".

--docdir=DIRECTORY
                 Sets the root directory for installing documentation
                 files, except "man" pages. This only sets the default for
                 the following options. The default value for this option
                 is "DATAROOTDIR/doc/postgresql".

--htmldir=DIRECTORY
                 The HTML-formatted documentation for PostgreSQL will be
                 installed under this directory. The default is
                 "DATAROOTDIR".

Note: Care has been taken to make it possible to install PostgreSQL
      into shared installation locations (such as "/usr/local/include")
      without interfering with the namespace of the rest of the system.
      First, the string "/postgresql" is automatically appended to
      datadir, sysconfdir, and docdir, unless the fully expanded directory
      name already contains the string "postgres" or "pgsql". For example,
      if you choose "/usr/local" as prefix, the documentation will be
      installed in "/usr/local/doc/postgresql", but if the prefix is
      "/opt/postgres", then it will be in "/opt/postgres/doc". The public
      C header files of the client interfaces are installed into
      includedir and are namespace-clean. The internal header files and
      the server header files are installed into private directories under
      includedir. 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 libdir for
      dynamically loadable modules.

--with-extra-version=STRING
                 Append "STRING" 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 "git describe"
                 identifier or a distribution package release number.

--with-includes=DIRECTORIES
                 "DIRECTORIES" 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
                 Readline) installed in a non-standard location, you have
                 to use this option and probably also the corresponding
                 "--with-libraries" option.

Example:
                 --with-includes=/opt/gnu/include:/usr/sup/include.

--with-libraries=DIRECTORIES
                 "DIRECTORIES" is a colon-separated list of directories to
                 search for libraries. You will probably have to use this
                 option (and the corresponding "--with-includes" option) if
                 you have packages installed in non-standard locations.

Example: --with-libraries=/opt/gnu/lib:/usr/sup/lib.

--enable-nls[=LANGUAGES]
                 Enables Native Language Support (NLS), that is, the
                 ability to display a program's messages in a language
                 other than English. "LANGUAGES" is an optional
                 space-separated list of codes of the languages that you
                 want supported, for example --enable-nls='de fr'. (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.

To use this option, you will need an implementation of the
                 Gettext API; see above.

--with-pgport=NUMBER
                 Set "NUMBER" 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 PostgreSQL servers on the same machine.

--with-perl
                 Build the PL/Perl server-side language.

--with-python
                 Build the PL/Python server-side language.

--with-tcl
                 Build the PL/Tcl server-side language.

--with-tclconfig=DIRECTORY
                 Tcl installs the file "tclConfig.sh", 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.

--with-gssapi
                 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., "/usr/include", "/usr/lib"),
                 so you must use the options "--with-includes" and
                 "--with-libraries" in addition to this option. "configure"
                 will check for the required header files and libraries to
                 make sure that your GSSAPI installation is sufficient
                 before proceeding.

--with-krb-srvnam=NAME
                 The default name of the Kerberos service principal used by
                 GSSAPI. postgres 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 POSTGRES.

--with-openssl
                 Build with support for SSL (encrypted) connections. This
                 requires the OpenSSL package to be installed. "configure"
                 will check for the required header files and libraries to
                 make sure that your OpenSSL installation is sufficient
                 before proceeding.

--with-pam
                 Build with PAM (Pluggable Authentication Modules) support.

--with-ldap
                 Build with LDAP support for authentication and connection
                 parameter lookup (see the documentation about client
                 authentication and libpq for more information). On Unix,
                 this requires the OpenLDAP package to be installed. On
                 Windows, the default WinLDAP library is used. "configure"
                 will check for the required header files and libraries to
                 make sure that your OpenLDAP installation is sufficient
                 before proceeding.

--without-readline
                 Prevents use of the Readline library (and libedit as
                 well). This option disables command-line editing and
                 history in psql, so it is not recommended.

--with-libedit-preferred
                 Favors the use of the BSD-licensed libedit library rather
                 than GPL-licensed Readline. This option is significant
                 only if you have both libraries installed; the default in
                 that case is to use Readline.

--with-bonjour
                 Build with Bonjour support. This requires Bonjour support
                 in your operating system. Recommended on OS X.

--with-uuid=LIBRARY
                 Build the uuid-ossp module (which provides functions to
                 generate UUIDs), using the specified UUID library.
                 "LIBRARY" must be one of:

o "bsd" to use the UUID functions found in FreeBSD, NetBSD,
                  and some other BSD-derived systems
                o "e2fs" to use the UUID library created by the e2fsprogs
                  project; this library is present in most Linux systems
                  and in OS X, and can be obtained for other platforms as
                  well
                o "ossp" to use the OSSP UUID library

--with-ossp-uuid
                 Obsolete equivalent of --with-uuid=ossp.

--with-libxml
                 Build with libxml (enables SQL/XML support). Libxml
                 version 2.6.23 or later is required for this feature.

Libxml installs a program "xml2-config" 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 XML2_CONFIG to point
                 to the "xml2-config" program belonging to the
                 installation, or use the options "--with-includes" and
                 "--with-libraries".

--with-libxslt
                 Use libxslt when building the xml2 module. xml2 relies on
                 this library to perform XSL transformations of XML.

--disable-integer-datetimes
                 Disable support for 64-bit integer storage for timestamps
                 and intervals, and store datetime values as floating-point
                 numbers instead. Floating-point datetime storage was the
                 default in PostgreSQL releases prior to 8.4, but it is now
                 deprecated, because it does not support microsecond
                 precision for the full range of timestamp values. However,
                 integer-based datetime storage requires a 64-bit integer
                 type. Therefore, this option can be used when no such type
                 is available, or for compatibility with applications
                 written for prior versions of PostgreSQL. See the
                 documentation about datetime datatypes for more
                 information.

--disable-float4-byval
                 Disable passing float4 values "by value", causing them to
                 be passed "by reference" instead. This option costs
                 performance, but may be needed for compatibility with old
                 user-defined functions that are written in C and use the
                 "version 0" calling convention. A better long-term
                 solution is to update any such functions to use the
                 "version 1" calling convention.

--disable-float8-byval
                 Disable passing float8 values "by value", causing them to
                 be passed "by reference" instead. This option costs
                 performance, but may be needed for compatibility with old
                 user-defined functions that are written in C and use the
                 "version 0" calling convention. A better long-term
                 solution is to update any such functions to use the
                 "version 1" calling convention. Note that this option
                 affects not only float8, but also int8 and some related
                 types such as timestamp. On 32-bit platforms,
                 "--disable-float8-byval" is the default and it is not
                 allowed to select "--enable-float8-byval".

--with-segsize=SEGSIZE
                 Set the segment size, 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 "largefile"
                 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
                 tar, 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.

--with-blocksize=BLOCKSIZE
                 Set the block size, 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.

--with-wal-segsize=SEGSIZE
                 Set the WAL segment size, in megabytes. This is the size
                 of each individual file in the WAL log. It may be useful
                 to adjust this size to control the granularity of WAL log
                 shipping. The default size is 16 megabytes. The value must
                 be a power of 2 between 1 and 64 (megabytes). Note that
                 changing this value requires an initdb.

--with-wal-blocksize=BLOCKSIZE
                 Set the WAL block size, 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.

--disable-spinlocks
                 Allow the build to succeed even if PostgreSQL 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 PostgreSQL on your platform,
                 please report the problem to the PostgreSQL developers.

--disable-thread-safety
                 Disable the thread-safety of client libraries. This
                 prevents concurrent threads in libpq and ECPG programs
                 from safely controlling their private connection handles.

--with-system-tzdata=DIRECTORY
                 PostgreSQL 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 "DIRECTORY" is used
                 instead of the one included in the PostgreSQL source
                 distribution. "DIRECTORY" must be specified as an absolute
                 path. "/usr/share/zoneinfo" 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 PostgreSQL.

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.

--without-zlib
                 Prevents use of the Zlib library. This disables support
                 for compressed archives in pg_dump and pg_restore. This
                 option is only intended for those rare systems where this
                 library is not available.

--enable-debug
                 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.

--enable-coverage
                 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. This option is for use only with GCC and when
                 doing development work.

--enable-profiling
                 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 "gmon.out" file for use in
                 profiling. This option is for use only with GCC and when
                 doing development work.

--enable-cassert
                 Enables assertion checks in the server, which test for
                 many "cannot happen" 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.

--enable-depend
                 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.

--enable-dtrace
                 Compiles PostgreSQL with support for the dynamic tracing
                 tool DTrace.

To point to the "dtrace" program, the environment variable
                 DTRACE can be set. This will often be necessary because
                 "dtrace" is typically installed under "/usr/sbin", which
                 might not be in the path.

Extra command-line options for the "dtrace" program can be
                 specified in the environment variable DTRACEFLAGS. On
                 Solaris, to include DTrace support in a 64-bit binary, you
                 must specify DTRACEFLAGS="-64" to configure. For example,
                 using the GCC compiler:

./configure CC='gcc -m64' --enable-dtrace DTRACEFLAGS='-64' ...

Using Sun's compiler:

./configure CC='/opt/SUNWspro/bin/cc -xtarget=native64' --enable-dtrace DTRACEFL

AGS='-64' ...

--enable-tap-tests
                 Enable tests using the Perl TAP tools. This requires a
                 Perl installation and the Perl module IPC::Run.

If you prefer a C compiler different from the one "configure"
        picks, you can set the environment variable CC to the program of
        your choice. By default, "configure" will pick "gcc" if available,
        else the platform's default (usually "cc"). Similarly, you can
        override the default compiler flags if needed with the CFLAGS
        variable.
        You can specify environment variables on the "configure" command
        line, for example:

./configure CC=/opt/bin/gcc CFLAGS='-O2 -pipe'
        Here is a list of the significant variables that can be set in this
        manner:

BISON
                 Bison program

CC
                 C compiler

CFLAGS
                 options to pass to the C compiler

CPP
                 C preprocessor

CPPFLAGS
                 options to pass to the C preprocessor

DTRACE
                 location of the "dtrace" program

DTRACEFLAGS
                 options to pass to the "dtrace" program

FLEX
                 Flex program

LDFLAGS
                 options to use when linking either executables or shared
                 libraries

LDFLAGS_EX
                 additional options for linking executables only

LDFLAGS_SL
                 additional options for linking shared libraries only

MSGFMT
                 "msgfmt" program for native language support

PERL
                 Full path to the Perl interpreter. This will be used to
                 determine the dependencies for building PL/Perl.

PYTHON
                 Full path to the Python interpreter. 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 the PL/Python
                 documentation for more information.

TCLSH
                 Full path to the Tcl interpreter. This will be used to
                 determine the dependencies for building PL/Tcl, and it
                 will be substituted into Tcl scripts.

XML2_CONFIG
                 "xml2-config" program used to locate the libxml
                 installation.

Note: When developing code inside the server, it is recommended to
      use the configure options "--enable-cassert" (which turns on many
      run-time error checks) and "--enable-debug" (which improves the
      usefulness of debugging tools).
      If using GCC, it is best to build with an optimization level of at
      least "-O1", because using no optimization ("-O0") 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 "-O0". An easy way to do this is by passing an
      option to make: "make PROFILE=-O0 file.o".
     2. Build
        To start the build, type:

make
        (Remember to use GNU make.) The build will take a few minutes
        depending on your hardware. The last line displayed should be:

All of PostgreSQL is successfully made. Ready to install.
        If you want to build everything that can be built, including the
        documentation (HTML and man pages), and the additional modules
        ("contrib"), type instead:

make world
        The last line displayed should be:

PostgreSQL, contrib and HTML documentation successfully made. Ready to install.
     3. Regression Tests
        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 PostgreSQL runs on your
        machine in the way the developers expected it to. Type:

make check
        (This won't work as root; do it as an unprivileged user.) The file
        "src/test/regress/README" and the documentation contain detailed
        information about interpreting the test results. You can repeat
        this test at any later time by issuing the same command.
     4. Installing the Files

Note: If you are upgrading an existing system be sure to read the
      documentation, which has instructions about upgrading a cluster.
        To install PostgreSQL enter:

make install
        This will install files into the directories that were specified in
        step 1. 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.
        To install the documentation (HTML and man pages), enter:

make install-docs
        If you built the world above, type instead:

make install-world
        This also installs the documentation.
        You can use make install-strip instead of make install 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. install-strip 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.
        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 PostgreSQL 8.0, a separate make install-all-headers
        command was needed for the latter, but this step has been folded
        into the standard install.)
        Client-only installation: If you want to install only the client
        applications and interface libraries, then you can use these
        commands:

make -C src/bin install

make -C src/include install

make -C src/interfaces install

make -C doc install
        "src/bin" has a few binaries for server-only use, but they are
        small.

Uninstallation: To undo the installation use the command "make
    uninstall". However, this will not remove any created directories.

Cleaning: After the installation you can free disk space by removing
    the built files from the source tree with the command "make clean".
    This will preserve the files made by the "configure" program, so that
    you can rebuild everything with "make" later on. To reset the source
    tree to the state in which it was distributed, use "make distclean". 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.)

If you perform a build and then discover that your "configure" options
    were wrong, or if you change anything that "configure" investigates
    (for example, software upgrades), then it's a good idea to do "make
    distclean" before reconfiguring and rebuilding. Without this, your
    changes in configuration choices might not propagate everywhere they
    need to.
      __________________________________________________________________

Post-Installation Setup

Shared Libraries

On some systems with shared libraries you need to tell the system how
    to find the newly installed shared libraries. The systems on which this
    is *not* necessary include FreeBSD, HP-UX, Linux, NetBSD, OpenBSD, and
    Solaris.

The method to set the shared library search path varies between
    platforms, but the most widely-used method is to set the environment
    variable LD_LIBRARY_PATH like so: In Bourne shells ("sh", "ksh",
    "bash", "zsh"):

LD_LIBRARY_PATH=/usr/local/pgsql/lib

export LD_LIBRARY_PATH

or in "csh" or "tcsh":

setenv LD_LIBRARY_PATH /usr/local/pgsql/lib

Replace /usr/local/pgsql/lib with whatever you set "--libdir" to in
    step 1. You should put these commands into a shell start-up file such
    as "/etc/profile" or "~/.bash_profile". Some good information about the
    caveats associated with this method can be found at
    http://xahlee.org/UnixResource_dir/_/ldpath.html.

On some systems it might be preferable to set the environment variable
    LD_RUN_PATH *before* building.

On Cygwin, put the library directory in the PATH or move the ".dll"
    files into the "bin" directory.

If in doubt, refer to the manual pages of your system (perhaps "ld.so"
    or "rld"). If you later get a message like:

psql: error in loading shared libraries

libpq.so.2.1: cannot open shared object file: No such file or directory

then this step was necessary. Simply take care of it then.

If you are on Linux and you have root access, you can run:

/sbin/ldconfig /usr/local/pgsql/lib

(or equivalent directory) after installation to enable the run-time
    linker to find the shared libraries faster. Refer to the manual page of
    "ldconfig" for more information. On FreeBSD, NetBSD, and OpenBSD the
    command is:

/sbin/ldconfig -m /usr/local/pgsql/lib

instead. Other systems are not known to have an equivalent command.
      __________________________________________________________________

Environment Variables

If you installed into "/usr/local/pgsql" or some other location that is
    not searched for programs by default, you should add
    "/usr/local/pgsql/bin" (or whatever you set "--bindir" to in step 1)
    into your PATH. Strictly speaking, this is not necessary, but it will
    make the use of PostgreSQL much more convenient.

To do this, add the following to your shell start-up file, such as
    "~/.bash_profile" (or "/etc/profile", if you want it to affect all
    users):

PATH=/usr/local/pgsql/bin:$PATH

export PATH

If you are using "csh" or "tcsh", then use this command:

set path = ( /usr/local/pgsql/bin $path )

To enable your system to find the man documentation, you need to add
    lines like the following to a shell start-up file unless you installed
    into a location that is searched by default:

MANPATH=/usr/local/pgsql/man:$MANPATH

export MANPATH

The environment variables PGHOST and PGPORT specify to client
    applications the host and port of the database server, overriding the
    compiled-in defaults. If you are going to run client applications
    remotely then it is convenient if every user that plans to use the
    database sets PGHOST. This is not required, however; the settings can
    be communicated via command line options to most client programs.
      __________________________________________________________________

Getting Started

The following is a quick summary of how to get PostgreSQL up and
    running once installed. The main documentation contains more
    information.
     1. Create a user account for the PostgreSQL server. This is the user
        the server will run as. For production use you should create a
        separate, unprivileged account ("postgres" is commonly used). If
        you do not have root access or just want to play around, your own
        user account is enough, but running the server as root is a
        security risk and will not work.

adduser postgres
     2. Create a database installation with the "initdb" command. To run
        "initdb" you must be logged in to your PostgreSQL server account.
        It will not work as root.

root# mkdir /usr/local/pgsql/data

root# chown postgres /usr/local/pgsql/data

root# su - postgres

postgres$ /usr/local/pgsql/bin/initdb -D /usr/local/pgsql/data
        The "-D" option specifies the location where the data will be
        stored. You can use any path you want, it does not have to be under
        the installation directory. Just make sure that the server account
        can write to the directory (or create it, if it doesn't already
        exist) before starting "initdb", as illustrated here.
     3. At this point, if you did not use the "initdb" -A option, you might
        want to modify "pg_hba.conf" to control local access to the server
        before you start it. The default is to trust all local users.
     4. The previous "initdb" step should have told you how to start up the
        database server. Do so now. The command should look something like:

/usr/local/pgsql/bin/postgres -D /usr/local/pgsql/data
        This will start the server in the foreground. To put the server in
        the background use something like:

nohup /usr/local/pgsql/bin/postgres -D /usr/local/pgsql/data \
     </dev/null >>server.log 2>&1 </dev/null &
        To stop a server running in the background you can type:

kill `cat /usr/local/pgsql/data/postmaster.pid`
     5. Create a database:

createdb testdb
        Then enter:

psql testdb
        to connect to that database. At the prompt you can enter SQL
        commands and start experimenting.
      __________________________________________________________________

What Now?

* The PostgreSQL distribution contains a comprehensive documentation
        set, which you should read sometime. After installation, the
        documentation can be accessed by pointing your browser to
        "/usr/local/pgsql/doc/html/index.html", unless you changed the
        installation directories.
        The first few chapters of the main documentation are the Tutorial,
        which should be your first reading if you are completely new to SQL
        databases. If you are familiar with database concepts then you want
        to proceed with part on server administration, which contains
        information about how to set up the database server, database
        users, and authentication.
      * Usually, you will want to modify your computer so that it will
        automatically start the database server whenever it boots. Some
        suggestions for this are in the documentation.
      * Run the regression tests against the installed server (using "make
        installcheck"). If you didn't run the tests before installation,
        you should definitely do it now. This is also explained in the
        documentation.
      * By default, PostgreSQL is configured to run on minimal hardware.
        This allows it to start up with almost any hardware configuration.
        The default configuration is, however, not designed for optimum
        performance. To achieve optimum performance, several server
        parameters must be adjusted, the two most common being
        shared_buffers and work_mem. Other parameters mentioned in the
        documentation also affect performance.
      __________________________________________________________________

Supported Platforms

A platform (that is, a CPU architecture and operating system
    combination) is considered supported by the PostgreSQL development
    community if the code contains provisions to work on that platform and
    it has recently been verified to build and pass its regression tests on
    that platform. Currently, most testing of platform compatibility is
    done automatically by test machines in the PostgreSQL Build Farm. If
    you are interested in using PostgreSQL on a platform that is not
    represented in the build farm, but on which the code works or can be
    made to work, you are strongly encouraged to set up a build farm member
    machine so that continued compatibility can be assured.

In general, PostgreSQL can be expected to work on these CPU
    architectures: x86, x86_64, IA64, PowerPC, PowerPC 64, S/390, S/390x,
    Sparc, Sparc 64, ARM, MIPS, MIPSEL, M68K, and PA-RISC. Code support
    exists for M32R and VAX, but these architectures are not known to have
    been tested recently. It is often possible to build on an unsupported
    CPU type by configuring with "--disable-spinlocks", but performance
    will be poor.

PostgreSQL can be expected to work on these operating systems: Linux
    (all recent distributions), Windows (Win2000 SP4 and later), FreeBSD,
    OpenBSD, NetBSD, OS X, AIX, HP/UX, Solaris, and UnixWare. Other
    Unix-like systems may also work but are not currently being tested. In
    most cases, all CPU architectures supported by a given operating system
    will work. Look in the the Section called Platform-specific Notes below
    to see if there is information specific to your operating system,
    particularly if using an older system.

If you have installation problems on a platform that is known to be
    supported according to recent build farm results, please report it to
    <pgsql-bugs@postgresql.org>. If you are interested in porting
    PostgreSQL to a new platform, <pgsql-hackers@postgresql.org> is the
    appropriate place to discuss that.
      __________________________________________________________________

Platform-specific Notes

This section documents additional platform-specific issues regarding
    the installation and setup of PostgreSQL. Be sure to read the
    installation instructions, and in particular the Section called
    Requirements as well. Also, check the file "src/test/regress/README"
    and the documentation regarding the interpretation of regression test
    results.

Platforms that are not covered here have no known platform-specific
    installation issues.
      __________________________________________________________________

AIX

PostgreSQL works on AIX, but getting it installed properly can be
    challenging. AIX versions from 4.3.3 to 6.1 are considered supported.
    You can use GCC or the native IBM compiler "xlc". In general, using
    recent versions of AIX and PostgreSQL helps. Check the build farm for
    up to date information about which versions of AIX are known to work.

The minimum recommended fix levels for supported AIX versions are:

AIX 4.3.3
           Maintenance Level 11 + post ML11 bundle

AIX 5.1
           Maintenance Level 9 + post ML9 bundle

AIX 5.2
           Technology Level 10 Service Pack 3

AIX 5.3
           Technology Level 7

AIX 6.1
           Base Level

To check your current fix level, use "oslevel -r" in AIX 4.3.3 to AIX
    5.2 ML 7, or "oslevel -s" in later versions.

Use the following "configure" flags in addition to your own if you have
    installed Readline or libz in /usr/local:
    --with-includes=/usr/local/include --with-libraries=/usr/local/lib.
      __________________________________________________________________

GCC Issues

On AIX 5.3, there have been some problems getting PostgreSQL to compile
    and run using GCC.

You will want to use a version of GCC subsequent to 3.3.2, particularly
    if you use a prepackaged version. We had good success with 4.0.1.
    Problems with earlier versions seem to have more to do with the way IBM
    packaged GCC than with actual issues with GCC, so that if you compile
    GCC yourself, you might well have success with an earlier version of
    GCC.
      __________________________________________________________________

Unix-Domain Sockets Broken

AIX 5.3 has a problem where sockaddr_storage is not defined to be large
    enough. In version 5.3, IBM increased the size of sockaddr_un, the
    address structure for Unix-domain sockets, but did not correspondingly
    increase the size of sockaddr_storage. The result of this is that
    attempts to use Unix-domain sockets with PostgreSQL lead to libpq
    overflowing the data structure. TCP/IP connections work OK, but not
    Unix-domain sockets, which prevents the regression tests from working.

The problem was reported to IBM, and is recorded as bug report
    PMR29657. If you upgrade to maintenance level 5300-03 or later, that
    will include this fix. A quick workaround is to alter _SS_MAXSIZE to
    1025 in "/usr/include/sys/socket.h". In either case, recompile
    PostgreSQL once you have the corrected header file.
      __________________________________________________________________

Internet Address Issues

PostgreSQL relies on the system's getaddrinfo function to parse IP
    addresses in listen_addresses, "pg_hba.conf", etc. Older versions of
    AIX have assorted bugs in this function. If you have problems related
    to these settings, updating to the appropriate AIX fix level shown
    above should take care of it.

One user reports:

When implementing PostgreSQL version 8.1 on AIX 5.3, we periodically
    ran into problems where the statistics collector would "mysteriously"
    not come up successfully. This appears to be the result of unexpected
    behavior in the IPv6 implementation. It looks like PostgreSQL and IPv6
    do not play very well together on AIX 5.3.

Any of the following actions "fix" the problem.

* Delete the IPv6 address for localhost:

(as root)

# ifconfig lo0 inet6 ::1/0 delete
      * Remove IPv6 from net services. The file "/etc/netsvc.conf" on AIX
        is roughly equivalent to "/etc/nsswitch.conf" on Solaris/Linux. The
        default, on AIX, is thus:

hosts=local,bind
        Replace this with:

hosts=local4,bind4
        to deactivate searching for IPv6 addresses.

Warning

This is really a workaround for problems relating to immaturity of IPv6
    support, which improved visibly during the course of AIX 5.3 releases.
    It has worked with AIX version 5.3, but does not represent an elegant
    solution to the problem. It has been reported that this workaround is
    not only unnecessary, but causes problems on AIX 6.1, where IPv6
    support has become more mature.
      __________________________________________________________________

Memory Management

AIX can be somewhat peculiar with regards to the way it does memory
    management. You can have a server with many multiples of gigabytes of
    RAM free, but still get out of memory or address space errors when
    running applications. One example is "createlang" failing with unusual
    errors. For example, running as the owner of the PostgreSQL
    installation:

-bash-3.00$ createlang plperl template1

createlang: language installation failed: ERROR:  could not load library "/opt/d

bs/pgsql748/lib/plperl.so": A memory address is not in the address space for the
  process.

Running as a non-owner in the group possessing the PostgreSQL
    installation:

-bash-3.00$ createlang plperl template1

createlang: language installation failed: ERROR:  could not load library "/opt/d

bs/pgsql748/lib/plperl.so": Bad address

Another example is out of memory errors in the PostgreSQL server logs,
    with every memory allocation near or greater than 256 MB failing.

The overall cause of all these problems is the default bittedness and
    memory model used by the server process. By default, all binaries built
    on AIX are 32-bit. This does not depend upon hardware type or kernel in
    use. These 32-bit processes are limited to 4 GB of memory laid out in
    256 MB segments using one of a few models. The default allows for less
    than 256 MB in the heap as it shares a single segment with the stack.

In the case of the "createlang" example, above, check your umask and
    the permissions of the binaries in your PostgreSQL installation. The
    binaries involved in that example were 32-bit and installed as mode 750
    instead of 755. Due to the permissions being set in this fashion, only
    the owner or a member of the possessing group can load the library.
    Since it isn't world-readable, the loader places the object into the
    process' heap instead of the shared library segments where it would
    otherwise be placed.

The "ideal" solution for this is to use a 64-bit build of PostgreSQL,
    but that is not always practical, because systems with 32-bit
    processors can build, but not run, 64-bit binaries.

If a 32-bit binary is desired, set LDR_CNTRL to MAXDATA=0xn0000000,
    where 1 <= n <= 8, before starting the PostgreSQL server, and try
    different values and "postgresql.conf" settings to find a configuration
    that works satisfactorily. This use of LDR_CNTRL tells AIX that you
    want the server to have MAXDATA bytes set aside for the heap, allocated
    in 256 MB segments. When you find a workable configuration, "ldedit"
    can be used to modify the binaries so that they default to using the
    desired heap size. PostgreSQL can also be rebuilt, passing configure
    LDFLAGS="-Wl,-bmaxdata:0xn0000000" to achieve the same effect.

For a 64-bit build, set OBJECT_MODE to 64 and pass CC="gcc -maix64" and
    LDFLAGS="-Wl,-bbigtoc" to "configure". (Options for "xlc" might
    differ.) If you omit the export of OBJECT_MODE, your build may fail
    with linker errors. When OBJECT_MODE is set, it tells AIX's build
    utilities such as "ar", "as", and "ld" what type of objects to default
    to handling.

By default, overcommit of paging space can happen. While we have not
    seen this occur, AIX will kill processes when it runs out of memory and
    the overcommit is accessed. The closest to this that we have seen is
    fork failing because the system decided that there was not enough
    memory for another process. Like many other parts of AIX, the paging
    space allocation method and out-of-memory kill is configurable on a
    system- or process-wide basis if this becomes a problem.

References and Resources

"Large Program Support", AIX Documentation: General Programming
    Concepts: Writing and Debugging Programs.

"Program Address Space Overview", AIX Documentation: General
    Programming Concepts: Writing and Debugging Programs.

"Performance Overview of the Virtual Memory Manager (VMM)", AIX
    Documentation: Performance Management Guide.

"Page Space Allocation", AIX Documentation: Performance Management
    Guide.

"Paging-space thresholds tuning", AIX Documentation: Performance
    Management Guide.

Developing and Porting C and C++ Applications on AIX, IBM Redbook.
      __________________________________________________________________

Cygwin

PostgreSQL can be built using Cygwin, a Linux-like environment for
    Windows, but that method is inferior to the native Windows build and
    running a server under Cygwin is no longer recommended.

When building from source, proceed according to the normal installation
    procedure (i.e., ./configure; make; etc.), noting the following-Cygwin
    specific differences:

* Set your path to use the Cygwin bin directory before the Windows
        utilities. This will help prevent problems with compilation.
      * The "adduser" command is not supported; use the appropriate user
        management application on Windows NT, 2000, or XP. Otherwise, skip
        this step.
      * The "su" command is not supported; use ssh to simulate su on
        Windows NT, 2000, or XP. Otherwise, skip this step.
      * OpenSSL is not supported.
      * Start "cygserver" for shared memory support. To do this, enter the
        command /usr/sbin/cygserver &. This program needs to be running
        anytime you start the PostgreSQL server or initialize a database
        cluster ("initdb"). The default "cygserver" configuration may need
        to be changed (e.g., increase SEMMNS) to prevent PostgreSQL from
        failing due to a lack of system resources.
      * Building might fail on some systems where a locale other than C is
        in use. To fix this, set the locale to C by doing "export
        LANG=C.utf8" before building, and then setting it back to the
        previous setting, after you have installed PostgreSQL.
      * The parallel regression tests (make check) can generate spurious
        regression test failures due to overflowing the listen() backlog
        queue which causes connection refused errors or hangs. You can
        limit the number of connections using the make variable
        MAX_CONNECTIONS thus:

make MAX_CONNECTIONS=5 check
        (On some systems you can have up to about 10 simultaneous
        connections).

It is possible to install "cygserver" and the PostgreSQL server as
    Windows NT services. For information on how to do this, please refer to
    the "README" document included with the PostgreSQL binary package on
    Cygwin. It is installed in the directory "/usr/share/doc/Cygwin".
      __________________________________________________________________

HP-UX

PostgreSQL 7.3+ should work on Series 700/800 PA-RISC machines running
    HP-UX 10.X or 11.X, given appropriate system patch levels and build
    tools. At least one developer routinely tests on HP-UX 10.20, and we
    have reports of successful installations on HP-UX 11.00 and 11.11.

Aside from the PostgreSQL source distribution, you will need GNU make
    (HP's make will not do), and either GCC or HP's full ANSI C compiler.
    If you intend to build from Git sources rather than a distribution
    tarball, you will also need Flex (GNU lex) and Bison (GNU yacc). We
    also recommend making sure you are fairly up-to-date on HP patches. At
    a minimum, if you are building 64 bit binaries on HP-UX 11.11 you may
    need PHSS_30966 (11.11) or a successor patch otherwise "initdb" may
    hang:

PHSS_30966  s700_800 ld(1) and linker tools cumulative patch
    On general principles you should be current on libc and ld/dld patches,
    as well as compiler patches if you are using HP's C compiler. See HP's
    support sites such as http://itrc.hp.com and
    ftp://us-ffs.external.hp.com/ for free copies of their latest patches.

If you are building on a PA-RISC 2.0 machine and want to have 64-bit
    binaries using GCC, you must use GCC 64-bit version. GCC binaries for
    HP-UX PA-RISC and Itanium are available from http://www.hp.com/go/gcc.
    Don't forget to get and install binutils at the same time.

If you are building on a PA-RISC 2.0 machine and want the compiled
    binaries to run on PA-RISC 1.1 machines you will need to specify
    "+DAportable" in CFLAGS.

If you are building on a HP-UX Itanium machine, you will need the
    latest HP ANSI C compiler with its dependent patch or successor
    patches:

PHSS_30848  s700_800 HP C Compiler (A.05.57)
    PHSS_30849  s700_800 u2comp/be/plugin library Patch

If you have both HP's C compiler and GCC's, then you might want to
    explicitly select the compiler to use when you run "configure":

./configure CC=cc

for HP's C compiler, or

./configure CC=gcc

for GCC. If you omit this setting, then configure will pick "gcc" if it
    has a choice.

The default install target location is "/usr/local/pgsql", which you
    might want to change to something under "/opt". If so, use the
    "--prefix" switch to "configure".

In the regression tests, there might be some low-order-digit
    differences in the geometry tests, which vary depending on which
    compiler and math library versions you use. Any other error is cause
    for suspicion.
      __________________________________________________________________

MinGW/Native Windows

PostgreSQL for Windows can be built using MinGW, a Unix-like build
    environment for Microsoft operating systems, or using Microsoft's
    Visual C++ compiler suite. The MinGW build variant uses the normal
    build system described in this chapter; the Visual C++ build works
    completely differently and is described in the documentation. It is a
    fully native build and uses no additional software like MinGW. A
    ready-made installer is available on the main PostgreSQL web site.

The native Windows port requires a 32 or 64-bit version of Windows 2000
    or later. Earlier operating systems do not have sufficient
    infrastructure (but Cygwin may be used on those). MinGW, the Unix-like
    build tools, and MSYS, a collection of Unix tools required to run shell
    scripts like "configure", can be downloaded from http://www.mingw.org/.
    Neither is required to run the resulting binaries; they are needed only
    for creating the binaries.

To build 64 bit binaries using MinGW, install the 64 bit tool set from
    http://mingw-w64.sourceforge.net/, put its bin directory in the PATH,
    and run "configure" with the "--host=x86_64-w64-mingw" option.

After you have everything installed, it is suggested that you run psql
    under "CMD.EXE", as the MSYS console has buffering issues.
      __________________________________________________________________

Collecting Crash Dumps on Windows

If PostgreSQL on Windows crashes, it has the ability to generate
    minidumps that can be used to track down the cause for the crash,
    similar to core dumps on Unix. These dumps can be read using the
    Windows Debugger Tools or using Visual Studio. To enable the generation
    of dumps on Windows, create a subdirectory named "crashdumps" inside
    the cluster data directory. The dumps will then be written into this
    directory with a unique name based on the identifier of the crashing
    process and the current time of the crash.
      __________________________________________________________________

SCO OpenServer and SCO UnixWare

PostgreSQL can be built on SCO UnixWare 7 and SCO OpenServer 5. On
    OpenServer, you can use either the OpenServer Development Kit or the
    Universal Development Kit. However, some tweaking may be needed, as
    described below.
      __________________________________________________________________

Skunkware

You should locate your copy of the SCO Skunkware CD. The Skunkware CD
    is included with UnixWare 7 and current versions of OpenServer 5.
    Skunkware includes ready-to-install versions of many popular programs
    that are available on the Internet. For example, gzip, gunzip, GNU
    Make, Flex, and Bison are all included. For UnixWare 7.1, this CD is
    now labeled "Open License Software Supplement". If you do not have this
    CD, the software on it is available from http://www.sco.com/skunkware/.

Skunkware has different versions for UnixWare and OpenServer. Make sure
    you install the correct version for your operating system, except as
    noted below.

On UnixWare 7.1.3 and beyond, the GCC compiler is included on the UDK
    CD as is GNU Make.
      __________________________________________________________________

GNU Make

You need to use the GNU Make program, which is on the Skunkware CD. By
    default, it installs as "/usr/local/bin/make".

As of UnixWare 7.1.3 and above, the GNU Make program is the OSTK
    portion of the UDK CD, and is in "/usr/gnu/bin/gmake".
      __________________________________________________________________

Readline

The Readline library is on the Skunkware CD. But it is not included on
    the UnixWare 7.1 Skunkware CD. If you have the UnixWare 7.0.0 or 7.0.1
    Skunkware CDs, you can install it from there. Otherwise, try
    http://www.sco.com/skunkware/.

By default, Readline installs into "/usr/local/lib" and
    "/usr/local/include". However, the PostgreSQL "configure" program will
    not find it there without help. If you installed Readline, then use the
    following options to "configure":

./configure --with-libraries=/usr/local/lib --with-includes=/usr/local/include
      __________________________________________________________________

Using the UDK on OpenServer

If you are using the new Universal Development Kit (UDK) compiler on
    OpenServer, you need to specify the locations of the UDK libraries:

./configure --with-libraries=/udk/usr/lib --with-includes=/udk/usr/include

Putting these together with the Readline options from above:

./configure --with-libraries="/udk/usr/lib /usr/local/lib" --with-includes="/udk

/usr/include /usr/local/include"
      __________________________________________________________________

Reading the PostgreSQL Man Pages

By default, the PostgreSQL man pages are installed into
    "/usr/local/pgsql/man". By default, UnixWare does not look there for
    man pages. To be able to read them you need to modify the MANPATH
    variable in "/etc/default/man", for example:

MANPATH=/usr/lib/scohelp/%L/man:/usr/dt/man:/usr/man:/usr/share/man:scohelp:/usr

/local/man:/usr/local/pgsql/man

On OpenServer, some extra research needs to be invested to make the man
    pages usable, because the man system is a bit different from other
    platforms. Currently, PostgreSQL will not install them at all.
      __________________________________________________________________

C99 Issues with the 7.1.1b Feature Supplement

For compilers earlier than the one released with OpenUNIX 8.0.0
    (UnixWare 7.1.2), including the 7.1.1b Feature Supplement, you may need
    to specify "-Xb" in CFLAGS or the CC environment variable. The
    indication of this is an error in compiling "tuplesort.c" referencing
    inline functions. Apparently there was a change in the 7.1.2(8.0.0)
    compiler and beyond.
      __________________________________________________________________

Threading on UnixWare

For threading, you*must* use "-Kpthread" on *all* libpq-using programs.
    libpq uses pthread_* calls, which are only available with the
    "-Kpthread"/"-Kthread" flag.
      __________________________________________________________________

Solaris

PostgreSQL is well-supported on Solaris. The more up to date your
    operating system, the fewer issues you will experience; details below.
      __________________________________________________________________

Required Tools

You can build with either GCC or Sun's compiler suite. For better code
    optimization, Sun's compiler is strongly recommended on the SPARC
    architecture. We have heard reports of problems when using GCC 2.95.1;
    GCC 2.95.3 or later is recommended. If you are using Sun's compiler, be
    careful not to select "/usr/ucb/cc"; use "/opt/SUNWspro/bin/cc".

You can download Sun Studio from
    http://www.oracle.com/technetwork/server-storage/solarisstudio/download
    s/. Many of GNU tools are integrated into Solaris 10, or they are
    present on the Solaris companion CD. If you like packages for older
    version of Solaris, you can find these tools at
    http://www.sunfreeware.com. If you prefer sources, look at
    http://www.gnu.org/order/ftp.html.
      __________________________________________________________________

Problems with OpenSSL

When you build PostgreSQL with OpenSSL support you might get
    compilation errors in the following files:

* "src/backend/libpq/crypt.c"
      * "src/backend/libpq/password.c"
      * "src/interfaces/libpq/fe-auth.c"
      * "src/interfaces/libpq/fe-connect.c"

This is because of a namespace conflict between the standard
    "/usr/include/crypt.h" header and the header files provided by OpenSSL.

Upgrading your OpenSSL installation to version 0.9.6a fixes this
    problem. Solaris 9 and above has a newer version of OpenSSL.
      __________________________________________________________________

configure Complains About a Failed Test Program

If "configure" complains about a failed test program, this is probably
    a case of the run-time linker being unable to find some library,
    probably libz, libreadline or some other non-standard library such as
    libssl. To point it to the right location, set the LDFLAGS environment
    variable on the "configure" command line, e.g.,

configure ... LDFLAGS="-R /usr/sfw/lib:/opt/sfw/lib:/usr/local/lib"

See the ld man page for more information.
      __________________________________________________________________

64-bit Build Sometimes Crashes

On Solaris 7 and older, the 64-bit version of libc has a buggy
    vsnprintf routine, which leads to erratic core dumps in PostgreSQL. The
    simplest known workaround is to force PostgreSQL to use its own version
    of vsnprintf rather than the library copy. To do this, after you run
    "configure" edit a file produced by "configure": In
    "src/Makefile.global", change the line

LIBOBJS =

to read

LIBOBJS = snprintf.o

(There might be other files already listed in this variable. Order does
    not matter.) Then build as usual.
      __________________________________________________________________

Compiling for Optimal Performance

On the SPARC architecture, Sun Studio is strongly recommended for
    compilation. Try using the "-xO5" optimization flag to generate
    significantly faster binaries. Do not use any flags that modify
    behavior of floating-point operations and errno processing (e.g.,
    "-fast"). These flags could raise some nonstandard PostgreSQL behavior
    for example in the date/time computing.

If you do not have a reason to use 64-bit binaries on SPARC, prefer the
    32-bit version. The 64-bit operations are slower and 64-bit binaries
    are slower than the 32-bit variants. And on other hand, 32-bit code on
    the AMD64 CPU family is not native, and that is why 32-bit code is
    significant slower on this CPU family.
      __________________________________________________________________

Using DTrace for Tracing PostgreSQL

Yes, using DTrace is possible. See the documentation for further
    information. You can also find more information in this article:
    https://blogs.oracle.com/robertlor/entry/user_level_dtrace_probes_in.

If you see the linking of the "postgres" executable abort with an error
    message like:

Undefined                       first referenced
  symbol                             in file

AbortTransaction                    utils/probes.o

CommitTransaction                   utils/probes.o

ld: fatal: Symbol referencing errors. No output written to postgres

collect2: ld returned 1 exit status

make: *** [postgres] Error 1

your DTrace installation is too old to handle probes in static
    functions. You need Solaris 10u4 or newer.