runtime.sgml

<!--
$Header: /cvsroot/pgsql/doc/src/sgml/runtime.sgml,v 1.59 2001/05/01 03:00:28 tgl Exp $
-->

<Chapter Id="runtime">
 <Title>Server Runtime Environment</Title>

 <Para>
  This chapter discusses how to set up and run the database server
  and the interactions with the operating system.
 </para>

 <sect1 id="postgres-user">
  <title>The Postgres user account</title>

  <para>
   As with any other server daemon that is connected to the world at
   large, it is advisable to run Postgres under a separate user
   account. This user account should only own the data itself that is
   being managed by the server, and should not be shared with other
   daemons. (Thus, using the user <quote>nobody</quote> is a bad
   idea.) It is not advisable to install the executables as owned by
   this user account because that runs the risk of user-defined
   functions gone astray or any other exploits compromising the
   executable programs.
  </para>

  <para>
   To add a user account to your system, look for a command
   <command>useradd</command> or <command>adduser</command>. The user
   name <quote>postgres</quote> is often used but by no means
   required.
  </para>
 </sect1>

 <sect1 id="creating-cluster">
  <title>Creating a database cluster</title>

  <para>
   Before you can do anything, you must initialize a database storage
   area on disk. We call this a <firstterm>database
   cluster</firstterm>. (<acronym>SQL</acronym> speaks of a catalog
   cluster instead.) A database cluster is a collection of databases
   that will be accessible through a single instance of a running
   database server. After initialization, a database cluster will
   contain one database named <literal>template1</literal>. As the
   name suggests, this will be used as a template for any subsequently
   created database; it should not be used for actual work.
  </para>

  <para>
   In file system terms, a database cluster will be a single directory
   under which all data will be stored. We call this the
   <firstterm>data directory</firstterm> or <firstterm>data
   area</firstterm>. It is completely up to you where you choose to
   store your data, there is no default, although locations such as
   <filename>/usr/local/pgsql/data</filename> or
   <filename>/var/lib/pgsql/data</filename> are popular. To initialize
   a database cluster, use the command <command>initdb</command>,
   which is installed with <productname>PostgreSQL</productname>. The
   desired file system location of your database system is indicated
   by the <option>-D</option> option, for example
<screen>
&gt; <userinput>initdb -D /usr/local/pgsql/data</userinput>
</screen>
   Note that you must execute this command while being logged in to
   the Postgres user account, which is described in the previous
   section.
  </para>

  <tip>
   <para>
    As an alternative to the <option>-D</option> option, you can set
    the environment variable <envar>PGDATA</envar>.
   </para>
  </tip>

  <para>
   <command>initdb</command> will attempt to create the directory you
   specify if it does not already exist. It is likely that it won't
   have the permission to do so (if you followed our advice and
   created an unprivileged account). In that case you should create the
   directory yourself (as root) and transfer ownership of it to the
   Postgres user account. Here is how this might work:
<screen>
root# <userinput>mkdir /usr/local/pgsql/data</userinput>
root# <userinput>chown postgres /usr/local/pgsql/data</userinput>
root# <userinput>su postgres</userinput>
postgres&gt; <userinput>initdb -D /usr/local/pgsql/data</userinput>
</screen>
  </para>

  <para>
   <command>initdb</command> will refuse to run if the data directory
   looks like it belongs to an already initialized installation.
  </para>

  <para>
   Because the data directory contains all the data stored in the
   database it is essential that it be well secured from unauthorized
   access. <command>initdb</command> therefore revokes access
   permissions from everyone but the Postgres user account.
  </para>

  <para>
   One surprise you might encounter while running <command>initdb</command> is
   a notice similar to this one:
<screen>
NOTICE:  Initializing database with en_US collation order.
        This locale setting will prevent use of index optimization for
        LIKE and regexp searches.  If you are concerned about speed of
        such queries, you may wish to set LC_COLLATE to "C" and
        re-initdb.  For more information see the Administrator's Guide.
</screen>
   This notice is intended to warn you that the currently selected locale
   will cause indexes to be sorted in an order that prevents them from
   being used for LIKE and regular-expression searches.  If you need
   good performance of such searches, you should set your current locale
   to "C" and re-run <command>initdb</command>.  On most systems, setting the
   current locale is done by changing the value of the environment variable
   <literal>LC_ALL</literal> or <literal>LANG</literal>.  The sort order used
   within a particular database cluster is set by <command>initdb</command>
   and cannot be changed later, short of dumping all data, re-initdb,
   reload data.  So it's important to make this choice correctly now.
  </para>
 </sect1>

 <sect1 id="postmaster-start">
  <title>Starting the database server</title>

  <para>
   Before anyone can access the database you must start the database
   server. The database server is called
   <firstterm>postmaster</firstterm>.
   The postmaster must know where to find the data it is supposed
   to work on. This is done with the <option>-D</option> option. Thus,
   the simplest way to start the server is, for example,
<screen>
&gt; <userinput>postmaster -D /usr/local/pgsql/data</userinput>
</screen>
   which will leave the server running in the foreground. This must
   again be done while logged in to the Postgres user account. Without
   a <option>-D</option>, the server will try to use the data
   directory in the environment variable <envar>PGDATA</envar>; if
   neither of these works it will fail.
  </para>

  <para>
   To start the <application>postmaster</application> in the
   background, use the usual shell syntax:
<screen>
&gt; <userinput>postmaster -D /usr/local/pgsql/data &gt; logfile 2&gt;&amp;1 &amp;</userinput>
</screen>
   It is an extremely good idea to keep the server output around
   somewhere, as indicated here. It will help both for auditing
   purposes and to diagnose problems.
  </para>

  <para>
   The postmaster also takes a number of other command line options.
   For more information see the reference page and below under runtime
   configuration. In particular, in order for the postmaster to accept
   TCP/IP connections (rather than just Unix domain socket ones), you
   must also specify the <option>-i</option> option.
  </para>

  <para>
   This shell syntax can get tedious quickly.  Therefore the shell
   script wrapper <application>pg_ctl</application> is provided that
   encapsulates some of the tasks.  E.g.,
<programlisting>
pg_ctl start -l logfile
</programlisting>
   will start the server in the background and put the output into the
   named log file.  The <option>-D</option> option has the same
   meaning as when invoking postmaster directly.
   <application>pg_ctl</application> also implements a symmetric
   <quote>stop</quote> operation.
  </para>

  <para>
   Normally, you will want to start the database server when the
   computer boots up. This is not required; the
   <productname>PostgreSQL</productname> server can be run
   successfully from non-privileged accounts without root
   intervention.
  </para>

  <para>
   Different systems have different conventions for starting up
   daemons at boot time, so you are advised to familiarize yourself
   with them. Many systems have a file
   <filename>/etc/rc.local</filename> or
   <filename>/etc/rc.d/rc.local</filename> which is almost certainly
   no bad place to put such a command. Whatever you do, the server
   must be run by the <productname>Postgres</productname> user account
   <emphasis>and not by root</emphasis> or any other user. Therefore
   you probably always want to form your command lines along the lines
   of <literal>su -c '...' postgres</literal>, for example:
<programlisting>
su -c 'pg_ctl -D /usr/local/pgsql/data -l serverlog' postgres
</programlisting>
  </para>

  <para>
   Here are a few more operating system specific suggestions.  (Always
   replace the proper installation directory and the user name you
   chose.)

   <itemizedlist>
    <listitem>
     <para>
      For <productname>FreeBSD</productname>, take a look at the file
      <filename>contrib/start-scripts/freebsd</filename> in the
      <productname>PostgreSQL</productname> source distribution.
     </para>
    </listitem>

    <listitem>
     <para>
      On <productname>OpenBSD</productname>, add the following lines
      to the file <filename>/etc/rc.local</filename>:
<programlisting>
if [ -x /usr/local/pgsql/bin/pg_ctl -a -x /usr/local/pgsql/bin/postmaster ]; then
    su - -c '/usr/local/pgsql/bin/pg_ctl start -l /var/postgresql/log -s' postgres
    echo -n ' postgresql'
fi
</programlisting>
     </para>
    </listitem>

    <listitem>
     <para>
      On <productname>Linux</productname> systems either add
<programlisting>
/usr/local/pgsql/bin/pg_ctl start -l logfile -D /usr/local/pgsql/data
</programlisting>
      to <filename>/etc/rc.d/rc.local</filename> or look into the file
      <filename>contrib/start-scripts/linux</filename> in the
      <productname>PostgreSQL</productname> source distribution to
      integrate the start and shutdown into the run level system.
     </para>
    </listitem>

    <listitem>
     <para>
      On <productname>NetBSD</productname>, either use the
      <productname>FreeBSD</productname> or
      <productname>Linux</productname> start scripts, depending on
      preference, as an example and place the file at
      <filename>/usr/local/etc/rc.d/postgresql</filename>.
     </para>
    </listitem>

    <listitem>
     <para>
      On <productname>Solaris</productname>, edit the file
      <filename>rc2.d</filename> to contain the following single line:
<programlisting>
su - postgres -c "/usr/local/pgsql/bin/pg_ctl start -l logfile -D /usr/local/pgsql/data"
</programlisting>
     </para>
    </listitem>
   </itemizedlist>

  </para>

   <para>
    While the <application>postmaster</application> is running, its
    PID is in the file <filename>postmaster.pid</filename> in the data
    directory. This is used as an interlock against multiple postmasters
    running in the same data directory, and can also be used for
    shutting down the postmaster.
   </para>

   <sect2 id="postmaster-start-failures">
    <title>Server Start-up Failures</title>

    <para>
     There are several common reasons for the postmaster to fail to
     start up. Check the postmaster's log file, or start it by hand
     (without redirecting standard output or standard error) to see
     what complaint messages appear. Some of the possible error
     messages are reasonably self-explanatory, but here are some that
     are not.
    </para>

    <para>
<screen>
FATAL: StreamServerPort: bind() failed: Address already in use
        Is another postmaster already running on that port?
</screen>
     This usually means just what it suggests: you accidentally
     started a second postmaster on the same port where one is already
     running. However, if the kernel error message is not
     <computeroutput>Address already in use</computeroutput> or some
     variant of that wording, there may be a different problem. For
     example, trying to start a postmaster on a reserved port number
     may draw something like
<screen>
&gt; <userinput>postmaster -i -p 666</userinput>
FATAL: StreamServerPort: bind() failed: Permission denied
        Is another postmaster already running on that port?
</screen>
    </para>

    <para>
     A message like
<screen>
IpcMemoryCreate: shmget(key=5440001, size=83918612, 01600) failed: Invalid argument
FATAL 1:  ShmemCreate: cannot create region
</screen>
     probably means that your kernel's limit on the size of shared
     memory areas is smaller than the buffer area that Postgres is
     trying to create (83918612 bytes in this example). Or it could
     mean that you don't have System-V-style shared memory support
     configured into your kernel at all. As a temporary workaround,
     you can try starting the postmaster with a smaller-than-normal
     number of buffers (<option>-B</option> switch). You will
     eventually want to reconfigure your kernel to increase the
     allowed shared memory size, however. You may see this message
     when trying to start multiple postmasters on the same machine, if
     their total space requests exceed the kernel limit.
    </para>

    <para>
     An error like
<screen>
IpcSemaphoreCreate: semget(key=5440026, num=16, 01600) failed: No space left on device
</screen>
     does <emphasis>not</emphasis> mean that you've run out of disk
     space; it means that your kernel's limit on the number of System
     V semaphores is smaller than the number
     <productname>Postgres</productname> wants to create. As above,
     you may be able to work around the problem by starting the
     postmaster with a reduced number of backend processes
     (<option>-N</option> switch), but you'll eventually want to
     increase the kernel limit.
    </para>

    <para>
     If you get an <quote>illegal system call</> error, then it is likely that
     shared memory or semaphores are not supported at all in your kernel. In
     that case your only option is to re-configure the kernel to turn on these
     features.
    </para>

    <para>
     Details about configuring System V IPC facilities are given in
     <xref linkend="sysvipc">.
    </para>
   </sect2>

   <sect2 id="client-connection-problems">
    <title>Client Connection Problems</title>

    <para>
     Although the possible error conditions on the client side are
     both virtually infinite and application dependent, a few of them
     might be directly related to how the server was started up.
     Conditions other than those shown below should be documented with
     the respective client application.
    </para>

    <para>
<screen>
PQconnectPoll() -- connect() failed: Connection refused
	Is the postmaster running (with -i) at 'server.joe.com'
	and accepting connections on TCP/IP port 5432?
</screen>
     This is the generic <quote>I couldn't find a server to talk
     to</quote> failure. It looks like the above when TCP/IP
     communication is attempted. A common mistake is to forget the
     <option>-i</option> to the postmaster to allow TCP/IP
     connections.
    </para>

    <para>
     Alternatively, you'll get this when attempting
     Unix-socket communication to a local postmaster:
<screen>
connectDBstart() -- connect() failed: No such file or directory
	Is the postmaster running locally
	and accepting connections on Unix socket '/tmp/.s.PGSQL.5432'?
</screen>
    </para>

    <para>
     The last line is useful in verifying that the client is trying to
     connect where it is supposed to. If there is in fact no
     postmaster running there, the kernel error message will typically
     be either <computeroutput>Connection refused</computeroutput> or
     <computeroutput>No such file or directory</computeroutput>, as
     illustrated. (It is particularly important to realize that
     <computeroutput>Connection refused</computeroutput> in this
     context does <emphasis>not</emphasis> mean that the postmaster
     got your connection request and rejected it -- that case will
     produce a different message, as shown in <xref
     linkend="client-authentication-problems">.) Other error messages
     such as <computeroutput>Connection timed out</computeroutput> may
     indicate more fundamental problems, like lack of network
     connectivity.
    </para>
   </sect2>
  </sect1>

  <sect1 id="runtime-config">
   <Title>Run-time configuration</Title>

   <para>
    There are a lot of configuration parameters that affect the
    behavior of the database system in some way or other. Here we
    describe how to set them and the following subsections will
    discuss each of them.
   </para>

   <para>
    All parameter names are case-insensitive. Every parameter takes a
    value of one of the four types boolean, integer, floating point,
    string as described below. Boolean values are
    <literal>ON</literal>, <literal>OFF</literal>,
    <literal>TRUE</literal>, <literal>FALSE</literal>,
    <literal>YES</literal>, <literal>NO</literal>,
    <literal>1</literal>, <literal>0</literal> (case-insensitive) or
    any non-ambiguous prefix of these.
   </para>

   <para>
    One way to set these options is to create a file
    <filename>postgresql.conf</filename> in the data directory (e.g.,
    <filename>/usr/local/pgsql/data</filename>). An example of what
    this file could look like is:
<programlisting>
# This is a comment
log_connections = yes
syslog = 2
</programlisting>
    As you see, options are one per line. The equal sign between name
    and value is optional. White space is insignificant, blank lines
    are ignored. Hash marks (<quote>#</quote>) introduce comments
    anywhere.
   </para>

   <para>
    The configuration file is reread whenever the postmaster receives
    a SIGHUP signal. This signal is also propagated to all running
    backend processes, so that running sessions get the new default.
    Alternatively, you can send the signal to only one backend process
    directly.
   </para>

   <para>
    A second way to set these configuration parameters is to give them
    as a command line option to the postmaster, such as
<programlisting>
postmaster -c log_connections=yes -c syslog=2
</programlisting>
    which would have the same effect as the previous example.
    Command-line options override any conflicting settings in
    <filename>postgresql.conf</filename>.
   </para>

   <para>
    Occasionally it is also useful to give a command line option to
    one particular backend session only. The environment variable
    <envar>PGOPTIONS</envar> can be used for this purpose on the
    client side:
<programlisting>
env PGOPTIONS='-c geqo=off' psql
</programlisting>
    (This works for any client application, not just
    <application>psql</application>.) Note that this won't work for
    options that are necessarily fixed once the server is started,
    such as the port number.
   </para>

   <para>
    Finally, some options can be changed in individual SQL sessions
    with the <command>SET</command> command, for example
<screen>
=&gt; <userinput>SET ENABLE_SEQSCAN TO OFF;</userinput>
</screen>
    See the SQL command language reference for details on the syntax.
   </para>

   <sect2 id="runtime-config-optimizer">
    <title>Planner and Optimizer Tuning</title>

   <para>
    <variablelist>
     <varlistentry>
      <term>CPU_INDEX_TUPLE_COST (<type>floating point</type>)</term>
      <listitem>
       <para>
        Sets the query optimizer's estimate of the cost of processing
	each index tuple during an index scan. This is measured as a
	fraction of the cost of a sequential page fetch.
       </para>
      </listitem>
     </varlistentry>
    
     <varlistentry>
      <term>CPU_OPERATOR_COST (<type>floating point</type>)</term>
      <listitem>
       <para>
        Sets the optimizer's estimate of the cost of processing each
	operator in a WHERE clause. This is measured as a fraction of
	the cost of a sequential page fetch.
       </para>
      </listitem>
     </varlistentry>
    
     <varlistentry>
      <term>CPU_TUPLE_COST (<type>floating point</type>)</term>
      <listitem>
       <para>
        Sets the query optimizer's estimate of the cost of processing
	each tuple during a query. This is measured as a fraction of
	the cost of a sequential page fetch.
       </para>
      </listitem>
     </varlistentry>
    
     <varlistentry>
      <term>EFFECTIVE_CACHE_SIZE (<type>floating point</type>)</term>
      <listitem>
       <para>
        Sets the optimizer's assumption about the effective size of
	the disk cache (that is, the portion of the kernel's disk
	cache that will be used for
	<productname>Postgres</productname> data files). This is
	measured in disk pages, which are normally 8kB apiece.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>ENABLE_HASHJOIN (<type>boolean</type>)</term>
      <listitem>
       <para>
        Enables or disables the query planner's use of hash-join plan
        types. The default is on. This is mostly useful to debug the
        query planner.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>ENABLE_INDEXSCAN (<type>boolean</type>)</term>
      <listitem>
       <para>
        Enables or disables the query planner's use of index scan plan
        types. The default is on. This is mostly useful to debug the
        query planner.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>ENABLE_MERGEJOIN (<type>boolean</type>)</term>
      <listitem>
       <para>
        Enables or disables the query planner's use of merge-join plan
        types. The default is on. This is mostly useful to debug the
        query planner.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>ENABLE_NESTLOOP (<type>boolean</type>)</term>
      <listitem>
       <para>
        Enables or disables the query planner's use of nested-loop
	join plans. It's not possible to suppress nested-loop joins
	entirely, but turning this variable off discourages the
	planner from using one if there is any other method available.
	The default is on. This is mostly useful to debug the query
	planner.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>ENABLE_SEQSCAN (<type>boolean</type>)</term>
      <listitem>
       <para>
        Enables or disables the query planner's use of sequential scan
	plan types. It's not possible to suppress sequential scans
	entirely, but turning this variable off discourages the
	planner from using one if there is any other method available.
	The default is on. This is mostly useful to debug the query
	planner.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>ENABLE_SORT (<type>boolean</type>)</term>
      <listitem>
       <para>
        Enables or disables the query planner's use of explicit sort
	steps. It's not possible to suppress explicit sorts entirely,
	but turning this variable off discourages the planner from
	using one if there is any other method available. The default
	is on. This is mostly useful to debug the query planner.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>ENABLE_TIDSCAN (<type>boolean</type>)</term>
      <listitem>
       <para>
        Enables or disables the query planner's use of TID scan plan
        types. The default is on. This is mostly useful to debug the
        query planner.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>GEQO (<type>boolean</type>)</term>
      <listitem>
       <para>
        Enables or disables genetic query optimization, which is an
        algorithm that attempts to do query planning without
        exhaustive search. This is on by default. See also the various
        other GEQO_ settings.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>GEQO_EFFORT (<type>integer</type>)</term>
      <term>GEQO_GENERATIONS (<type>integer</type>)</term>
      <term>GEQO_POOL_SIZE (<type>integer</type>)</term>
      <term>GEQO_RANDOM_SEED (<type>integer</type>)</term>
      <term>GEQO_SELECTION_BIAS (<type>floating point</type>)</term>
      <listitem>
       <para>
        Various tuning parameters for the genetic query optimization
        algorithm: The pool size is the number of individuals in one
        population. Valid values are between 128 and 1024. If it is
        set to 0 (the default) a pool size of 2^(QS+1), where QS
        is the number of FROM items in the query, is taken. The effort
        is used to calculate a default for generations. Valid values
        are between 1 and 80, 40 being the default. Generations
        specifies the number of iterations in the algorithm. The
        number must be a positive integer. If 0 is specified then
        Effort * Log2(PoolSize) is used. The run time of the algorithm
        is roughly proportional to the sum of pool size and
        generations. The selection bias is the selective pressure
        within the population. Values can be from 1.50 to 2.00; the
        latter is the default. The random seed can be set to get
        reproduceable results from the algorithm. If it is set to -1
        then the algorithm behaves non-deterministically.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>GEQO_THRESHOLD (<type>integer</type>)</term>
      <listitem>
       <para>
        Use genetic query optimization to plan queries with at least
        this many FROM items involved.  (Note that a JOIN construct
	counts as only one FROM item.) The default is 11. For simpler
	queries it is usually best to use the
        deterministic, exhaustive planner.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>KSQO (<type>boolean</type>)</term>
      <listitem>
       <para>
        The <firstterm>Key Set Query Optimizer</firstterm>
        (<abbrev>KSQO</abbrev>) causes the query planner to convert
        queries whose WHERE clause contains many OR'ed AND clauses
        (such as <literal>WHERE (a=1 AND b=2) OR (a=2 AND b=3)
        ...</literal>) into a UNION query. This method can be faster
        than the default implementation, but it doesn't necessarily
        give exactly the same results, since UNION implicitly adds a
        SELECT DISTINCT clause to eliminate identical output rows.
        KSQO is commonly used when working with products like
        <productname>Microsoft Access</productname>, which tend to
        generate queries of this form.
       </para>

       <para>
        The KSQO algorithm used to be absolutely essential for queries
        with many OR'ed AND clauses, but in
        <productname>Postgres</productname> 7.0 and later the standard
        planner handles these queries fairly successfully. Hence the
        default is OFF.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>RANDOM_PAGE_COST (<type>floating point</type>)</term>
      <listitem>
       <para>
        Sets the query optimizer's estimate of the cost of a
	nonsequentially fetched disk page. This is measured as a
	multiple of the cost of a sequential page fetch.
       </para>
      </listitem>
     </varlistentry>
    </variablelist>
   </para>

   <note>
    <para>
     Unfortunately, there is no well-defined method of determining
     ideal values for the family of <quote>COST</quote> variables that
     were just described. You are encouraged to experiment and share
     your findings.
    </para>
   </note>

   </sect2>

   <sect2 id="logging">
    <title>Logging and Debugging</title>

   <para>
    <variablelist>
     <varlistentry>
      <term>DEBUG_ASSERTIONS (<type>boolean</type>)</term>
      <listitem>
       <para>
        Turns on various assertion checks. This is a debugging aid. If
        you are experiencing strange problems or crashes you might
        want to turn this on, as it might expose programming mistakes.
        To use this option, the macro <literal>USE_ASSERT_CHECKING</literal>
        must be defined when Postgres is built (see the configure option
	<literal>--enable-cassert</literal>).  Note that
	<literal>DEBUG_ASSERTIONS</literal> defaults to ON if Postgres
	has been built this way.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>DEBUG_LEVEL (<type>integer</type>)</term>
      <listitem>
       <para>
        The higher this value is set, the more
        <quote>debugging</quote> output of various sorts is generated
        in the server log during operation. This option is 0 by
        default, which means no debugging output. Values up to about 4
        currently make sense.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>DEBUG_PRINT_QUERY (<type>boolean</type>)</term>
      <term>DEBUG_PRINT_PARSE (<type>boolean</type>)</term>
      <term>DEBUG_PRINT_REWRITTEN (<type>boolean</type>)</term>
      <term>DEBUG_PRINT_PLAN (<type>boolean</type>)</term>
      <term>DEBUG_PRETTY_PRINT (<type>boolean</type>)</term>
      <listitem>
       <para>
        These flags enable various debugging output to be sent to the
	server log.  For each executed query, prints either the query text,
	the resulting parse tree, the query rewriter output, or the execution
	plan. <option>DEBUG_PRETTY_PRINT</option> indents these displays
	to produce a more readable but much longer output format.
	Setting <option>DEBUG_LEVEL</option> above zero implicitly turns
	on some of these flags.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>HOSTNAME_LOOKUP (<type>boolean</type>)</term>
      <listitem>
       <para>
        By default, connection logs only show the IP address of the
        connecting host. If you want it to show the host name you can
        turn this on, but depending on your host name resolution setup
        it might impose a non-negligible performance penalty. This
        option can only be set at server start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>LOG_CONNECTIONS (<type>boolean</type>)</term>
      <listitem>
       <para>
        Prints a line informing about each successful connection to
        the server log. This is off by default, although it is
        probably very useful. This option can only be set at server
        start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>LOG_PID (<type>boolean</type>)</term>
      <listitem>
       <para>
        Prefixes each server log message with the process id of the
        backend process. This is useful to sort out which messages
        pertain to which connection. The default is off.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>LOG_TIMESTAMP (<type>boolean</type>)</term>
      <listitem>
       <para>
        Prefixes each server log message with a timestamp. The default
        is off.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>SHOW_QUERY_STATS (<type>boolean</type>)</term>
      <term>SHOW_PARSER_STATS (<type>boolean</type>)</term>
      <term>SHOW_PLANNER_STATS (<type>boolean</type>)</term>
      <term>SHOW_EXECUTOR_STATS (<type>boolean</type>)</term>
      <listitem>
       <para>
        For each query, write performance statistics of the respective
        module to the server log. This is a crude profiling
        instrument.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>SHOW_SOURCE_PORT (<type>boolean</type>)</term>
      <listitem>
       <para>
        Shows the outgoing port number of the connecting host in the
        connection log messages. You could trace back the port number
        to find out what user initiated the connection. Other than
        that it's pretty useless and therefore off by default. This
        option can only be set at server start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>SYSLOG (<type>integer</type>)</term>
      <listitem>
       <para>
        <productname>Postgres</productname> allows the use of
        <application>syslog</application> for logging. If this option
        is set to 1, messages go both to syslog and the standard
        output. A setting of 2 sends output only to syslog. (Some
        messages will still go to the standard output/error.) The
        default is 0, which means syslog is off. This option must be
        set at server start.
       </para>
       <para>
        To use syslog, the build of
        <productname>Postgres</productname> must be configured with
        the <option>--enable-syslog</option> option.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>SYSLOG_FACILITY (<type>string</type>)</term>
       <listitem>
        <para>
          This option determines the <application>syslog</application>
          <quote>facility</quote> to be used when syslog is enabled.
          You may choose from LOCAL0, LOCAL1, LOCAL2, LOCAL3, LOCAL4,
          LOCAL5, LOCAL6, LOCAL7; the default is LOCAL0.  See also the
          documentation of your system's
          <application>syslog</application>.
        </para>
       </listitem>
     </varlistentry>
     
     <varlistentry>
      <term>SYSLOG_IDENT (<type>string</type>)</term>
       <listitem>
        <para>
         If logging to syslog is enabled, this option determines the
         program name used to identify
         <productname>PostgreSQL</productname> messages in
         <application>syslog</application> log messages.  The default
         is <quote>postgres</quote>.
        </para>
       </listitem>
      </varlistentry>

     <varlistentry>
      <term>TRACE_NOTIFY (<type>boolean</type>)</term>
      <listitem>
       <para>
        Generates a great amount of debugging output for the
        <command>LISTEN</command> and <command>NOTIFY</command>
        commands.
       </para>
      </listitem>
     </varlistentry>
    </variablelist>
   </para>
   </sect2>

   <sect2 id="runtime-config-general">
    <title>General operation</title>

   <para>
    <variablelist>
     <varlistentry>
      <term>DEADLOCK_TIMEOUT (<type>integer</type>)</term>
      <listitem>
       <para>
        This is the amount of time, in milliseconds, to wait on a lock
	before checking to see if there is a deadlock condition or not.
	The check for deadlock is relatively slow, so we don't want to
	run it every time we wait for a lock.  We (optimistically?)
	assume that deadlocks are not common in production applications,
	and just wait on the lock for awhile before starting to ask
	questions about whether it can ever get unlocked.
	Increasing this value reduces the amount of time wasted in
	needless deadlock checks, but slows down reporting of real deadlock
	errors.  The default is 1000 (i.e., one second), which is probably
	about the smallest value you would want in practice.  On a heavily
	loaded server you might want to raise it.  Ideally the setting
	should exceed your typical transaction time, so as to improve the
	odds that the lock will be released before the waiter decides to
	check for deadlock. 
	This option can only be set at server start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>FSYNC (<type>boolean</type>)</term>
      <listitem>
       <para>
        If this option is on, the <productname>Postgres</> backend
        will use the <function>fsync()</> system call in several
        places to make sure that updates are physically written to
        disk and do not hang around in the kernel buffer cache. This
        increases the chance that a database installation will still
        be usable after an operating system or hardware crash by a
        large amount. (Crashes of the database server itself do
        <emphasis>not</> affect this consideration.)
       </para>

       <para>
        However, this operation slows down <productname>Postgres</>,
	because at all those points it has
        to block and wait for the operating system to flush the
        buffers. Without <function>fsync</>, the operating system is
        allowed to do its best in buffering, sorting, and delaying
        writes, which can make for a considerable perfomance
        increase. However, if the system crashes, the results of the
	last few committed transactions may be lost in part or whole;
	in the worst case, unrecoverable data corruption may occur.
       </para>

       <para>
        This option is the subject of an eternal debate in the
        <productname>Postgres</> user and developer communities. Some
        always leave it off, some turn it off only for bulk loads,
        where there is a clear restart point if something goes wrong,
        some leave it on just to be on the safe side. Because it is
        the safe side, on is also the default. If you trust your
        operating system, your hardware, and your utility company (or
	better your UPS), you might want to disable fsync.
       </para>

       <para>
        It should be noted that the performance penalty from doing
	fsyncs is considerably less in <productname>Postgres</> version
	7.1 than it was in prior releases.  If you previously suppressed
	fsyncs because of performance problems, you may wish to reconsider
	your choice.
       </para>

       <para>
	This option can only be set at server start or in the
	<filename>postgresql.conf</filename> file.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>KRB_SERVER_KEYFILE (<type>string</type>)</term>
      <listitem>
       <para>
        Sets the location of the Kerberos server key file. See
        <xref linkend="kerberos-auth"> for details.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>MAX_CONNECTIONS (<type>integer</type>)</term>
      <listitem>
       <para>
        Determines how many concurrent connections the database server
        will allow. The default is 32. There is also a compiled-in
        hard upper limit on this value, which is typically 1024
	(both numbers can be altered when compiling the server). This
        parameter can only be set at server start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>MAX_EXPR_DEPTH (<type>integer</type>)</term>
      <listitem>
       <para>
        Sets the maximum expression nesting depth that the parser will
	accept. The default value is high enough for any normal query,
	but you can raise it if you need to. (But if you raise it too
	high, you run the risk of backend crashes due to stack
	overflow.)
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>PORT (<type>integer</type>)</term>
      <listitem>
       <para>
        The TCP port the server listens on; 5432 by default. This
        option can only be set at server start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>SHARED_BUFFERS (<type>integer</type>)</term>
      <listitem>
       <para>
        Sets the number of shared memory buffers the database server
        will use. The default is 64. Each buffer is typically 8192
        bytes. This option can only be set at server start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>SILENT_MODE (<type>bool</type>)</term>
      <listitem>
       <para>
        Runs postmaster silently. If this option is set, postmaster
        will automatically run in background and any controlling ttys
        are disassociated, thus no messages are written to stdout or
        stderr (same effect as postmaster's -S option). Unless some
        logging system such as syslog is enabled, using this option is
        discouraged since it makes it impossible to see error
        messages.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>SORT_MEM (<type>integer</type>)</term>
      <listitem>
       <para>
	Specifies the amount of memory to be used by internal sorts
	and hashes before resorting to temporary disk files. The value
	is specified in kilobytes, and defaults to 512 kilobytes. Note
	that for a complex query, several sorts and/or hashes might be
	running in parallel, and each one will be allowed to use as
	much memory as this value specifies before it starts to put
	data into temporary files.  And don't forget that each running
	backend could be doing one or more sorts.  So the total memory
	space needed could be many times the value of SORT_MEM.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>SQL_INHERITANCE (<type>bool</type>)</term>
      <listitem>
       <para>
        This controls the inheritance semantics, in particular whether
        subtables are included into the consideration of various
        commands by default. This was not the case in versions prior
        to 7.1. If you need the old behaviour you can set this
        variable to off, but in the long run you are encouraged to
        change your applications to use the <literal>ONLY</literal>
        keyword to exclude subtables. See the SQL language reference
        and the <citetitle>User's Guide</citetitle> for more
        information about inheritance.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>SSL (<type>boolean</type>)</term>
      <listitem>
       <para>
        Enables <acronym>SSL</> connections. Please read
        <xref linkend="ssl-tcp"> before using this. The default
        is off.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>TCPIP_SOCKET (<type>boolean</type>)</term>
      <listitem>
       <para>
        If this is true, then the server will accept TCP/IP
        connections. Otherwise only local Unix domain socket
        connections are accepted. It is off by default. This option
        can only be set at server start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>UNIX_SOCKET_DIRECTORY (<type>string</type>)</term>
      <listitem>
       <para>
	Specifies the directory of the Unix-domain socket on which the
	<application>postmaster</application> is to listen for
	connections from client applications.  The default is normally
	<filename>/tmp</filename>, but can be changed at build time.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>UNIX_SOCKET_GROUP (<type>string</type>)</term>
      <listitem>
       <para>
        Sets the group owner of the Unix domain socket.  (The owning
        user of the socket is always the user that starts the
        postmaster.)  In combination with the option
        <option>UNIX_SOCKET_PERMISSIONS</option> this can be used as
        an additional access control mechanism for this socket type.
        By default this is the empty string, which uses the default
        group for the current user.  This option can only be set at
        server start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>UNIX_SOCKET_PERMISSIONS (<type>integer</type>)</term>
      <listitem>
       <para>
        Sets the access permissions of the Unix domain socket.  Unix
        domain sockets use the usual Unix file system permission set.
        The option value is expected to be an numeric mode
        specification in the form accepted by the
        <function>chmod</function> and <function>umask</function>
        system calls.  (To use the customary octal format the number
        must start with a <literal>0</literal> (zero).)
       </para>

       <para>
        The default permissions are <literal>0777</literal>, meaning
        anyone can connect.  Reasonable alternatives would be
        <literal>0770</literal> (only user and group, see also under
        <option>UNIX_SOCKET_GROUP</option>) and
        <literal>0700</literal> (only user).  (Note that actually for
        a Unix socket, only write permission matters and there is no
        point in setting or revoking read or execute permissions.)
       </para>

       <para>
        This access control mechanism is independent from the one
        described in <xref linkend="client-authentication">.
       </para>

       <para>
        This option can only be set at server start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>VIRTUAL_HOST (<type>string</type>)</term>
      <listitem>
       <para>
	Specifies the TCP/IP hostname or address on which the
	<application>postmaster</application> is to listen for
	connections from client applications.  Defaults to
	listening on all configured addresses (including localhost).
       </para>
      </listitem>
     </varlistentry>

    </variablelist>
   </para>
   </sect2>

   <sect2 id="runtime-config-wal">
    <title>WAL</title>

   <para>
    See also <xref linkend="wal-configuration"> for details on WAL
    tuning.

    <variablelist>
     <varlistentry>
      <term>CHECKPOINT_SEGMENTS (<type>integer</type>)</term>
      <listitem>
       <para>
        Maximum distance between automatic WAL checkpoints, in logfile
	segments (each segment is normally 16 megabytes).
	This option can only be set at server start or in the
	<filename>postgresql.conf</filename> file.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>CHECKPOINT_TIMEOUT (<type>integer</type>)</term>
      <listitem>
       <para>
        Maximum time between automatic WAL checkpoints, in seconds.
	This option can only be set at server start or in the
	<filename>postgresql.conf</filename> file.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>WAL_BUFFERS (<type>integer</type>)</term>
      <listitem>
       <para>
        Number of disk-page buffers in shared memory for WAL log.
	This option can only be set at server start.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>WAL_DEBUG (<type>integer</type>)</term>
      <listitem>
       <para>
        If non-zero, turn on WAL-related debugging output on standard
        error.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>WAL_FILES (<type>integer</type>)</term>
      <listitem>
       <para>
        Number of log files that are created in advance at checkpoint
        time.  This option can only be set at server start or in the
	<filename>postgresql.conf</filename> file.
       </para>
      </listitem>
     </varlistentry>

     <varlistentry>
      <term>WAL_SYNC_METHOD (<type>string</type>)</term>
      <listitem>
       <para>
        Method used for forcing WAL updates out to disk.  Possible
	values are
	<literal>FSYNC</> (call fsync() at each commit),
	<literal>FDATASYNC</> (call fdatasync() at each commit),
	<literal>OPEN_SYNC</> (write WAL files with open() option O_SYNC), or
	<literal>OPEN_DATASYNC</> (write WAL files with open() option O_DSYNC).
	Not all of these choices are available on all platforms.
        This option can only be set at server start or in the
	<filename>postgresql.conf</filename> file.
       </para>
      </listitem>
     </varlistentry>
    </variablelist>
    </para>
   </sect2>


   <sect2 id="runtime-config-short">
    <title>Short options</title>
   <para>
    For convenience there are also single letter option switches
    available for many parameters. They are described in the following
    table.

    <table>
     <title>Short option key</title>
     <tgroup cols="3">
      <colspec colnum="3" align="center">
      <thead>
       <row>
        <entry>Short option</entry>
        <entry>Equivalent</entry>
        <entry>Remark</entry>
       </row>
      </thead>
      <tbody>
       <row>
        <entry>-B <replaceable>x</replaceable></entry>
        <entry>shared_buffers = <replaceable>x</replaceable></entry>
        <entry></entry>
       </row>
       <row>
        <entry>-d <replaceable>x</replaceable></entry>
        <entry>debug_level = <replaceable>x</replaceable></entry>
        <entry></entry>
       </row>
       <row>
        <entry>-F</entry>
        <entry>fsync = off</entry>
        <entry></entry>
       </row>
       <row>
        <entry>-h <replaceable>x</replaceable></entry>
        <entry>virtual_host = <replaceable>x</replaceable></entry>
        <entry></entry>
       </row>
       <row>
        <entry>-i</entry>
        <entry>tcpip_socket = on</entry>
        <entry></entry>
       </row>
       <row>
        <entry>-k <replaceable>x</replaceable></entry>
        <entry>unix_socket_directory = <replaceable>x</replaceable></entry>
        <entry></entry>
       </row>
       <row>
        <entry>-l</entry>
        <entry>ssl = on</entry>
        <entry></entry>
       </row>
       <row>
        <entry>-N <replaceable>x</replaceable></entry>
        <entry>max_connections = <replaceable>x</replaceable></entry>
        <entry></entry>
       </row>
       <row>
        <entry>-p <replaceable>x</replaceable></entry>
        <entry>port = <replaceable>x</replaceable></entry>
        <entry></entry>
       </row>

       <row>
        <entry>-fi, -fh, -fm, -fn, -fs, -ft</entry>
        <entry>enable_indexscan=off, enable_hashjoin=off,
        enable_mergejoin=off, enable_nestloop=off, enable_seqscan=off,
        enable_tidscan=off</entry>
        <entry>*</entry>
       </row>
       <row>
        <entry>-S <replaceable>x</replaceable></entry>
        <entry>sort_mem = <replaceable>x</replaceable></entry>
        <entry>*</entry>
       </row>
       <row>
        <entry>-s</entry>
        <entry>show_query_stats = on</entry>
        <entry>*</entry>
       </row>
       <row>
        <entry>-tpa, -tpl, -te</entry>
        <entry>show_parser_stats=on, show_planner_stats=on, show_executor_stats=on</entry>
        <entry>*</entry>
       </row>
      </tbody>
     </tgroup>
    </table>
    For historical reasons, options marked <quote>*</quote> must be
    passed to the individual backend process via the
    <option>-o</option> postmaster option, for example,
<screen>
&gt; <userinput>postmaster -o '-S 1024 -s'</userinput>
</screen>
    or via <envar>PGOPTIONS</envar> from the client side, as explained
    above.
   </para>

   </sect2>
 </sect1>


 <sect1 id="kernel-resources">
  <title>Managing Kernel Resources</title>

  <para>
   A large <productname>Postgres</> installation can quickly hit
   various operating system resource limits. (On some systems, the
   factory defaults are so low that you don't even need a really
   <quote>large</> installation.) If you have encountered this kind of
   problem then keep reading.
  </para>

  <sect2 id="sysvipc">
   <title>Shared Memory and Semaphores</title>

   <para>
    Shared memory and semaphores are collectively referred to as
    <quote>System V IPC</> (together with message queues, which are
    not relevant for <productname>Postgres</>). Almost all modern
    operating systems provide these features, but not all of them have
    them turned on or sufficiently sized by default, especially
    systems with BSD heritage. (For the QNX and BeOS ports,
    <productname>Postgres</> provides its own replacement
    implementation of these facilities.)
   </para>

   <para>
    The complete lack of these facilities is usually manifested by an
    <errorname>Illegal system call</> error upon postmaster start. In
    that case there's nothing left to do but to reconfigure your
    kernel -- <productname>Postgres</> won't work without them.
   </para>

   <para>
    When <productname>Postgres</> exceeds one of the various hard
    limits of the IPC resources then the postmaster will refuse to
    start up and should leave a marginally instructive error message
    about which problem was encountered and what needs to be done
    about it.  (See also <xref linkend="postmaster-start-failures">.)
    The relevant kernel parameters are named
    consistently across different systems; <xref
    linkend="sysvipc-parameters"> gives an overview. The methods to
    set them, however, vary; suggestions for some platforms are given
    below. Be warned that it is often necessary to reboot your
    machine at least, possibly even recompile the kernel, to change these
    settings.
   </para>


   <table id="sysvipc-parameters">
    <title>System V IPC parameters</>

    <tgroup cols="3">
     <thead>
      <row>
       <entry>Name</>
       <entry>Description</>
       <entry>Reasonable values</>
      </row>
     </thead>

     <tbody>
      <row>
       <entry><varname>SHMMAX</></>
       <entry>Maximum size of shared memory segment (bytes)</>
       <entry>512 kB + 8192 * buffers + extra ... infinity</entry>
      </row>

      <row>
       <entry><varname>SHMMIN</></>
       <entry>Minimum size of shared memory segment (bytes)</>
       <entry>1 (at most about 256 kB)</>
      </row>

      <row>
       <entry><varname>SHMSEG</></>
       <entry>Maximum number of shared memory segments per process</>
       <entry>only 1 segment is needed, but the default is much higher</>
      </row>

       <row>
        <entry><varname>SHMMNI</></>
        <entry>Maximum number of shared memory segments system-wide</>
        <entry>like <varname>SHMSEG</> + room for other applications</>
       </row>

       <row>
        <entry><varname>SEMMNI</></>
        <entry>Maximum number of semaphore identifiers (i.e., sets)</>
        <entry>&gt;= ceil(max_connections / 16)</>
       </row>

       <row>
        <entry><varname>SEMMNS</></>
        <entry>Maximum number of semaphores system-wide</>
        <entry>ceil(max_connections / 16) * 17 + room for other applications</>
       </row>

       <row>
        <entry><varname>SEMMSL</></>
        <entry>Maximum number of semaphores per set</>
        <entry>&gt;= 17</>
       </row>

       <row>
        <entry><varname>SEMMAP</></>
        <entry>Number of entries in semaphore map</>
        <entry>see text</>
       </row>

       <row>
        <entry><varname>SEMVMX</></>
        <entry>Maximum value of semaphore</>
        <entry>&gt;= 255  (The default is often 32767, don't change unless asked to.)</>
       </row>

     </tbody>
    </tgroup>
   </table>


   <para>
    The most important shared memory parameter is <varname>SHMMAX</>,
    the maximum size, in bytes, that a shared memory segment can have.
    If you get an error message from <function>shmget</> along the
    lines of <errorname>Invalid argument</> then it is possible that
    this limit has been exceeded. The size of the required shared
    memory segments varies both with the number of requested buffers
    (<option>-B</> option) and the number of allowed connections
    (<option>-N</> option), although the former is the dominant item.
    (You can therefore, as a temporary solution, lower these settings
    to get rid of the failures.) As a rough approximation you can
    estimate the required segment size as the number of buffers times
    the block size (8192 kB by default) plus ample overhead (at least
    half a megabyte). Any error message you might get will contain the
    size of the failed allocation request.
   </para>

   <para>
    Less likely to cause problems is the minimum size for shared
    memory segments (<varname>SHMMIN</>), which should be at most
    somewhere around 256 kB for <productname>Postgres</> (it is
    usually just 1). The maximum number of segments system-wide
    (<varname>SHMMNI</>) or per-process (<varname>SHMSEG</>) should
    not cause a problem unless your system has them set to zero. Some
    systems also have a limit on the total amount of shared memory in
    the system; see the platform-specific instructions below.
   </para>

   <para>
    <productname>Postgres</> uses one semaphore per allowed connection
    (<option>-N</> option), in sets of 16.  Each such set will also
    contain a 17th semaphore which contains a <quote>magic
    number</quote>, to avoid collision with semaphore sets used by
    other applications. The maximum number of semaphores in the system
    is set by <varname>SEMMNS</>, which consequently must be at least
    as high as the connection setting plus one extra for each 16
    allowed connections (see the formula in <xref
    linkend="sysvipc-parameters">).  The parameter <varname>SEMMNI</>
    determines the limit on the number of semaphore sets that can
    exist on the system at one time.  Hence this parameter must be at
    least <literal>ceil(max_connections / 16)</>. Lowering the number
    of allowed connections is a temporary workaround for failures,
    which are usually confusingly worded <quote><errorname>No space
    left on device</></>, from the function <function>semget()</>.
   </para>

   <para>
    In some cases it might also turn out to be necessary to increase
    <varname>SEMMAP</> to be at least on the order of
    <varname>SEMMNS</>. This parameter defines the size of the
    semaphore resource map, in which each contiguous block of available
    semaphores needs an entry. When a semaphore set is freed it is
    either added to an existing entry that is adjacent to the freed
    block or it is registered under a new map entry. If the map is
    full, the freed semaphores get lost (until reboot). Fragmentation
    of the semaphore space could therefore over time lead to less
    available semaphores than there should be.
   </para>

   <para>
    The <varname>SEMMSL</> parameter, which determines how many
    semaphores can be in a set, must be at least 17 for
    <productname>Postgres</>.
   </para>

   <para>
    Various other settings related to <quote>semaphore undo</>, such as
    <varname>SEMMNU</> and <varname>SEMUME</>, are not of concern
    for <productname>Postgres</>.
   </para>


   <para>
    <variablelist>

     <varlistentry>
      <term>BSD/OS</>
      <listitem>
       <formalpara>
        <title>Shared Memory</>
        <para>
         By default, only 4 MB of shared memory is supported. Keep in
         mind that shared memory is not pageable; it is locked in RAM.
	 The shared memory parameters are:
<programlisting>
#define SHMMAX		/* max shared memory segment size (bytes) */
#define SHMMIN		/* min shared memory segment size (bytes) */
#define SHMMNI		/* max number of shared memory identifiers */
#define SHMSEG		/* max shared memory segments per process */
#define SHMALL		/* max amount of shared memory (pages) */
</programlisting>

	 To increase the number of buffers supported by the postmaster, add the
	 following to your kernel config file.  A <varname>SHMALL</> value of 1024 
	 represents 4MB of shared memory.  Increase it accordingly:
<programlisting>
options "SHMALL=4096"
options "SHMMAX=\(SHMALL*PAGE_SIZE\)"
</programlisting>
        </para>
       </formalpara>

       <para>
        For those running 4.1 or later, just recompile the kernel and
        reboot. For those running earlier releases, use
        <application>bpatch</> to find the <varname>sysptsize</> value
        for the current kernel. This is computed dynamically at
        bootup.
<screen>
$ <userinput>bpatch -r sysptsize</>
<computeroutput>0x9 = 9</>
</screen>
        Next, change <varname>SYSPTSIZE</> to a hard-coded value. Use
        the bpatch value, plus add 1 for every additional 4 MB of
        shared memory you desire.
<programlisting>
options "SYSPTSIZE=13"
</programlisting>
        <varname>sysptsize</> can not be changed by sysctl on the fly.
       </para>

       <formalpara>
        <title>Semaphores</>
        <para>
         You may need to increase the number of semaphores. By
         default, <productname>Postgres</> allocates 34 semaphores,
	 which is over half the default system total of 60.
        </para>
       </formalpara>

       <para>
        The defaults are in <filename>/sys/sys/sem.h</>:
<programlisting>
#define SEMMNI  10              /* # of semaphore identifiers */

#define SEMMNS  60              /* # of semaphores in system */

#define SEMUME  10              /* max # of undo entries per process */

#define SEMMNU  30              /* # of undo structures in system */
</programlisting>
        Set the values you want in your kernel config file, e.g.:
<programlisting>
options "SEMMNI=40"
options "SEMMNS=240"
options "SEMUME=40"
options "SEMMNU=120"
</programlisting>
       </para>
      </listitem>
     </varlistentry>


     <varlistentry>
      <term>FreeBSD</term>
      <term>OpenBSD</term>
      <listitem>
       <para>
        The options <varname>SYSVSHM</> and <varname>SYSVSEM</> need
        to be enabled when the kernel is compiled. (They are by
        default.) The maximum size of shared memory is determined by
        the option <varname>SHMMAXPGS</> (in pages). The following
        shows an example of how to set the various parameters:
<programlisting>
options         SYSVSHM
options         SHMMAXPGS=4096
options         SHMSEG=256

options         SYSVSEM
options         SEMMNI=256
options         SEMMNS=512
options         SEMMNU=256
options         SEMMAP=256
</programlisting>
       </para>
      </listitem>
     </varlistentry>


     <varlistentry>
      <term>HP-UX</>
      <listitem>
       <para>
        The default settings tend to suffice for normal installations.
        On <productname>HP-UX</> 10, the factory default for
        <varname>SEMMNS</> is 128, which might be too low for larger
        database sites.
       </para>
       <para>
        IPC parameters can be set in the <application>System
        Administration Manager</> (<acronym>SAM</>) under
        <menuchoice><guimenu>Kernel
        Configuration</><guimenuitem>Configurable Parameters</></>.
        Hit <guibutton>Create A New Kernel</> when you're done.
       </para>
      </listitem>
     </varlistentry>


     <varlistentry>
      <term>Linux</>
      <listitem>
       <para>
        The default shared memory limit (both
        <varname>SHMMAX</varname> and <varname>SHMALL</varname>) is 32
        MB in 2.2 kernels, but it can be changed in the
        <filename>proc</filename> file system (without reboot).  For
        example, to allow 128 MB:
<screen>
<prompt>$</prompt> <userinput>echo 134217728 >/proc/sys/kernel/shmall</userinput>
<prompt>$</prompt> <userinput>echo 134217728 >/proc/sys/kernel/shmmax</userinput>
</screen>
        You could put these commands into a script run at boot-time.
       </para>

       <para>
        Alternatively, you can use
        <citerefentry><refentrytitle>sysctl</refentrytitle>
        <manvolnum>8</manvolnum></citerefentry>, if available, to
        control these parameters.  Look for a file called
        <filename>/etc/sysctl.conf</filename> and add lines like the
        following to it:
<programlisting>
kernel.shmall = 134217728
kernel.shmmax = 134217728
</programlisting>
        This file is usually processed at boot time, but
        <application>sysctl</application> can also be called
        explicitly later.
       </para>

       <para>
        Other parameters are sufficiently sized for any application.
        If you want to see for yourself look into
        <filename>/usr/src/linux/include/asm-<replaceable>xxx</>/shmparam.h</>
        and <filename>/usr/src/linux/include/linux/sem.h</>.
       </para>
      </listitem>
     </varlistentry>


     <varlistentry>
      <term>SCO OpenServer</>
      <listitem>
       <para>
        In the default configuration, only 512 kB of shared memory per
        segment is allowed, which is about enough for <option>-B 24 -N
        12</>. To increase the setting, first change the directory to
        <filename>/etc/conf/cf.d</>. To display the current value of
        <varname>SHMMAX</>, in bytes, run
<programlisting>
./configure -y SHMMAX
</programlisting>
        To set a new value for <varname>SHMMAX</>, run:
<programlisting>
./configure SHMMAX=<replaceable>value</>
</programlisting>
        where <replaceable>value</> is the new value you want to use
        (in bytes). After setting <varname>SHMMAX</>, rebuild the kernel
<programlisting>
./link_unix
</programlisting>
        and reboot.
       </para>
      </listitem>
     </varlistentry>


     <varlistentry>
      <term>Solaris</>
      <listitem>
       <para>
        At least in version 2.6, the maximum size of a shared memory
        segment is set too low for <productname>Postgres</>. The
        relevant settings can be changed in <filename>/etc/system</>,
        for example:
<programlisting>
set shmsys:shminfo_shmmax=0x2000000
set shmsys:shminfo_shmmin=1
set shmsys:shminfo_shmmni=256
set shmsys:shminfo_shmseg=256

set semsys:seminfo_semmap=256
set semsys:seminfo_semmni=512
set semsys:seminfo_semmns=512
set semsys:seminfo_semmsl=32
</programlisting>
        You need to reboot to make the changes effective.
       </para>

       <para>
        See also <ulink
        url="http://www.sunworld.com/swol-09-1997/swol-09-insidesolaris.html">http://www.sunworld.com/swol-09-1997/swol-09-insidesolaris.html</>
        for information on shared memory under
        <productname>Solaris</>.
       </para>
      </listitem>
     </varlistentry>


     <varlistentry>
      <term>UnixWare</>
      <listitem>
       <para>
        On <productname>UnixWare</> 7, the maximum size for shared
        memory segments is 512 kB in the default configuration. This
        is enough for about <option>-B 24 -N 12</>. To display the
        current value of <varname>SHMMAX</>, run
<programlisting>
/etc/conf/bin/idtune -g SHMMAX
</programlisting>
        which displays the current, default, minimum, and maximum
        values, in bytes. To set a new value for <varname>SHMMAX</>,
        run:
<programlisting>
/etc/conf/bin/idtune SHMMAX <replaceable>value</>
</programlisting>
        where <replaceable>value</> is the new value you want to use
        (in bytes). After setting <varname>SHMMAX</>, rebuild the
        kernel
<programlisting>
/etc/conf/bin/idbuild -B
</programlisting>
        and reboot.
       </para>
      </listitem>
     </varlistentry>

    </variablelist>

   </para>
  </sect2>


  <sect2>
   <title>Resource Limits</title>

   <para>
    Unix-like operating systems enforce various kinds of resource
    limits that might interfere with the operation of your
    <productname>Postgres</productname> server.  Of importance are
    especially the limits on the number of processes per user, the
    number of open files per process, and the amount of memory
    available to a process.  Each of these have a <quote>hard</quote>
    and a <quote>soft</quote> limit.  The soft limit is what actually
    counts but it can be changed by the user up to the hard limit.
    The hard limit can only be changed by the root user.  The system
    call <function>setrlimit</function> is responsible for setting
    these parameters.  The shell's built-in command
    <command>ulimit</command> (Bourne shells) or
    <command>limit</command> (csh) is used to control the resource
    limits from the command line.  On BSD-derived systems the file
    <filename>/etc/login.conf</filename> controls what values the
    various resource limits are set to upon login.  See
    <citerefentry><refentrytitle>login.conf</refentrytitle>
    <manvolnum>5</manvolnum></citerefentry> for details.  The relevant
    parameters are <varname>maxproc</varname>,
    <varname>openfiles</varname>, and <varname>datasize</varname>.
    For example:
<programlisting>
default:\
...
        :datasize-cur=256M:\
        :maxproc-cur=256:\
        :openfiles-cur=256:\
...
</programlisting>
    (<literal>-cur</literal> is the soft limit.  Append
    <literal>-max</literal> to set the hard limit.)
   </para>

   <para>
    Kernels generally also have an implementation-dependent
    system-wide limit on some resources.
    <itemizedlist>
     <listitem>
      <para>
      On <productname>Linux</productname>
      <filename>/proc/sys/fs/file-max</filename> determines the
      maximum number of files that the kernel will allocate.  It can
      be changed by writing a different number into the file or by
      adding an assignment in <filename>/etc/sysctl.conf</filename>.
      The maximum limit of files per process is fixed at the time the
      kernel is compiled; see
      <filename>/usr/src/linux/Documentation/proc.txt</filename> for
      more information.
      </para>
     </listitem>
    </itemizedlist>
   </para>

   <para>
    The <productname>Postgres</productname> server uses one process
    per connection so you should provide for at least as many processes
    as allowed connections, in addition to what you need for the rest
    of your system.  This is usually not a problem but if you run
    several servers on one machine things might get tight.
   </para>

   <para>
    The factory default limit on open files is often set to
    <quote>socially friendly</quote> values that allow many users to
    coexist on a machine without using an inappropriate fraction of
    the system resources.  If you run many servers on a machine this
    is perhaps what you want, but on dedicated servers you may want to
    raise this limit.
   </para>
  </sect2>

 </sect1>


 <sect1 id="postmaster-shutdown">
  <title>Shutting down the server</title>

  <para>
   Depending on your needs, there are several ways to shut down the
   database server when your work is done. The differentiation is
   done by what signal you send to the server process.
   <variablelist>
    <varlistentry>
     <term>SIGTERM</term>
     <listitem>
      <para>
       After receiving SIGTERM, the postmaster disallows new
       connections, but lets existing backends end their work normally.
       It shuts down only after all of the backends terminate by client
       request.
       This is the <firstterm>Smart Shutdown</firstterm>.
      </para>
     </listitem>
    </varlistentry>

    <varlistentry>
     <term>SIGINT</term>
     <listitem>
      <para>
       The postmaster disallows new connections and sends all existing
       backends SIGTERM, which will cause them to abort their current
       transactions and exit promptly.  It then waits for the backends to exit
       and finally shuts down the data base.
       This is the <firstterm>Fast Shutdown</firstterm>.
      </para>
     </listitem>
    </varlistentry>

    <varlistentry>
     <term>SIGQUIT</term>
     <listitem>
      <para>
       This is the <firstterm>Immediate Shutdown</firstterm> which
       will cause the postmaster to send a SIGQUIT to all backends and
       exit immediately (without properly shutting down the database
       system).  The backends likewise exit immediately upon receiving
       SIGQUIT. This will lead to recovery (by replaying the WAL log)
       upon next start-up.  This is recommended only in emergencies.
      </para>
     </listitem>
    </varlistentry>
   </variablelist>

   <caution>
    <para>
     It is best not to use SIGKILL to shut down the postmaster. This
     will prevent the postmaster from releasing shared memory and
     semaphores, which you may then have to do by hand.
    </para>
   </caution>

   The PID of the postmaster process can be found using the
   <application>ps</application> program, or from the file
   <filename>postmaster.pid</filename> in the data directory. So for
   example, to do a fast shutdown:
<screen>
&gt; <userinput>kill -INT `head -1 /usr/local/pgsql/data/postmaster.pid`</userinput>
</screen>
  </para>
  <para>
   The program <application>pg_ctl</application> is a shell script
   that provides a more convenient interface for shutting down the
   postmaster.
  </para>
 </sect1>

 <sect1 id="ssl-tcp">
  <title>Secure TCP/IP Connections with SSL</title>

  <para>
   <productname>PostgreSQL</> has native support for connections over
   <acronym>SSL</> to encrypt
   client/server communications for increased security. This requires
   <productname>OpenSSL</productname> to be installed on both client
   and server systems and support enabled at build-time (see <xref
   linkend="installation">).
  </para>

  <para>
   With SSL support compiled in, the <productname>PostgreSQL</> server
   can be started with the argument <option>-l</> (ell) to enable
   SSL connections. When starting in SSL mode, the postmaster will look
   for the files <filename>server.key</> and <filename>server.crt</> in
   the data directory.  These files should contain the server private key
   and certificate respectively. These files must be set up correctly
   before an SSL-enabled server can start. If the private key is protected
   with a passphrase, the postmaster will prompt for the passphrase and will
   not start until it has been entered.
  </para>

  <para>
   The postmaster will listen for both standard and SSL connections
   on the same TCP/IP port, and will negotiate with any connecting
   client whether or not to use SSL.
   See <xref linkend="client-authentication">
   about how to force on the server side the use of SSL for certain
   connections.
  </para>

  <para>
   For details on how to create your server private key and certificate,
   refer to the <productname>OpenSSL</> documentation. A simple self-signed
   certificate can be used to get started for testing, but a certificate signed
   by a CA (either one of the global CAs or a local one) should be used in 
   production so the client can verify the servers identity. To create
   a quick self-signed certificate, use the following OpenSSL command:
<programlisting>
openssl req -new -text -out cert.req
</programlisting>
   Fill out the information that openssl asks for. Make sure that you enter
   the local host name as Common Name; the challenge password can be
   left blank. The script will generate a key that is passphrase protected;
   it will not accept a pass phrase that is less than four characters long.
   To remove the passphrase (as you must if you want automatic start-up of
   the postmaster), run the commands
<programlisting>
openssl rsa -in privkey.pem -out cert.pem
</programlisting>
   Enter the old passphrase to unlock the existing key. Now do
<programlisting>
openssl req -x509 -in cert.req -text -key cert.pem -out cert.cert
cp cert.pem <replaceable>$PGDATA</replaceable>/server.key
cp cert.cert <replaceable>$PGDATA</replaceable>/server.crt
</programlisting>
   to turn the certificate into a self-signed certificate and to copy the
   key and certificate to where the postmaster will look for them.
  </para>
 </sect1>

 <sect1 id="ssh-tunnels">
  <title>Secure TCP/IP Connections with SSH tunnels</title>

  <note>
   <title>Acknowledgement</title>
   <para>
    Idea taken from an email by Gene Selkov, Jr.
    (<email>selkovjr@mcs.anl.gov</>) written on 1999-09-08 in response
    to a question from Eric Marsden.
   </para>
  </note>

  <para>
   One can use <productname>ssh</productname> to encrypt the network
   connection between clients and a
   <productname>Postgres</productname> server. Done properly, this
   should lead to an adequately secure network connection.
  </para>

  <para>
   First make sure that an <productname>ssh</productname> server is
   running properly on the same machine as
   <productname>Postgres</productname> and that you can log in using
   ssh as some user. Then you can establish a secure tunnel with a
   command like this from the client machine:
<programlisting>
&gt; <userinput>ssh -L 3333:foo.com:5432 joe@foo.com</userinput>
</programlisting>
   The first number in the <option>-L</option> argument, 3333, is the
   port number of your end of the tunnel; it can be chosen freely. The
   second number, 5432, is the remote end of the tunnel -- the port
   number your backend is using. The name or the address in between
   the port numbers is the host with the database server you are going
   to connect to. In order to connect to the database server using
   this tunnel, you connect to port 3333 on the local machine:
<programlisting>
psql -h localhost -p 3333 template1
</programlisting>
   To the database server it will then look as though you are really
   user <literal>joe@foo.com</literal> and it will use whatever
   authentication procedure was set up for this user. In order for the
   tunnel setup to succeed you must be allowed to connect via ssh as
   joe@foo.com, just as if you had attempted to use ssh to set up a
   terminal session.
  </para>

  <tip>
   <para>
    Several other products exist that can provide secure tunnels using
    a procedure similar in concept to the one just described.
   </para>
  </tip>

 </sect1>

</Chapter>

<!-- Keep this comment at the end of the file
Local variables:
mode:sgml
sgml-omittag:nil
sgml-shorttag:t
sgml-minimize-attributes:nil
sgml-always-quote-attributes:t
sgml-indent-step:1
sgml-indent-data:t
sgml-parent-document:nil
sgml-default-dtd-file:"./reference.ced"
sgml-exposed-tags:nil
sgml-local-catalogs:("/usr/lib/sgml/catalog")
sgml-local-ecat-files:nil
End:
-->