[postgresql] / doc / src / sgml / pgupgrade.sgml

<!-- doc/src/sgml/pgupgrade.sgml -->

<sect1 id="pgupgrade" xreflabel="pg_upgrade">
 <title>pg_upgrade</title>

 <indexterm zone="pgupgrade">
  <primary>pg_upgrade</primary>
 </indexterm>

 <para>
  <application>pg_upgrade</> (formerly called <application>pg_migrator</>) allows data
  stored in <productname>PostgreSQL</> data files to be upgraded to a later <productname>PostgreSQL</>
  major version without the data dump/reload typically required for
  major version upgrades, e.g. from 8.4.7 to the current major release
  of <productname>PostgreSQL</>.  It is not required for minor version upgrades, e.g. from
  9.0.1 to 9.0.4.
 </para>

 <para>
  Major PostgreSQL releases regularly add new features that often
  change the layout of the system tables, but the internal data storage
  format rarely changes.  <application>pg_upgrade</> uses this fact
  to perform rapid upgrades by creating new system tables and simply
  reusing the old user data files.  If a future major release ever
  changes the data storage format in a way that makes the old data
  format unreadable, <application>pg_upgrade</> will not be usable
  for such upgrades.  (The community will attempt to avoid such
  situations.)
 </para>

 <para>
  <application>pg_upgrade</> does its best to
  make sure the old and new clusters are binary-compatible, e.g.  by
  checking for compatible compile-time settings, including 32/64-bit
  binaries.  It is important that
  any external modules are also binary compatible, though this cannot
  be checked by <application>pg_upgrade</>.
 </para>

 <sect2>
  <title>Supported Versions</title>

  <para>
   pg_upgrade supports upgrades from 8.3.X and later to the current
   major release of <productname>PostgreSQL</>, including snapshot and alpha releases.

  </para>

 </sect2>

 <sect2>
  <title><application>pg_upgrade</> Options</title>

   <para>
    <application>pg_upgrade</application> accepts the following command-line arguments:

    <variablelist>

     <varlistentry>
      <term><option>-b</option> <replaceable>old_bindir</></term>
      <term><option>--old-bindir=</option><replaceable>old_bindir</></term>
      <listitem><para>the old cluster executable directory;
      environment variable <envar>OLDBINDIR</></para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-B</option> <replaceable>new_bindir</></term>
      <term><option>--new-bindir=</option><replaceable>new_bindir</></term>
      <listitem><para>the new cluster executable directory;
      environment variable <envar>NEWBINDIR</></para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-c</option></term>
      <term><option>--check</option></term>
      <listitem><para>check clusters only, don't change any data</para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-d</option> <replaceable>old_datadir</></term>
      <term><option>--old-datadir=</option><replaceable>old_datadir</></term>
      <listitem><para>the old cluster data directory; environment
      variable <envar>OLDDATADIR</></para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-D</option> <replaceable>new_datadir</></term>
      <term><option>--new-datadir=</option><replaceable>new_datadir</></term>
      <listitem><para>the new cluster data directory; environment
      variable <envar>NEWDATADIR</></para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-g</option></term>
      <term><option>--debug</option></term>
      <listitem><para>enable debugging</para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-G</option> <replaceable>debug_filename</></term>
      <term><option>--debugfile=</option><replaceable>debug_filename</></term>
      <listitem><para>output debugging activity to file</para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-k</option></term>
      <term><option>--link</option></term>
      <listitem><para>use hard links instead of copying files to the new cluster</para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-l</option> <replaceable>log_filename</></term>
      <term><option>--logfile=</option><replaceable>log_filename</></term>
      <listitem><para>log session activity to file</para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-p</option> <replaceable>old_port_number</></term>
      <term><option>--old-port=</option><replaceable>old_portnum</></term>
      <listitem><para>the old cluster port number; environment
      variable <envar>PGPORT</></para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-P</option> <replaceable>new_port_number</></term>
      <term><option>--new-port=</option><replaceable>new_portnum</></term>
      <listitem><para>the new cluster port number; environment
      variable <envar>PGPORT</></para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-u</option> <replaceable>user_name</></term>
      <term><option>--user=</option><replaceable>user_name</></term>
      <listitem><para>cluster's super user name; environment
      variable <envar>PGUSER</></para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-v</option></term>
      <term><option>--verbose</option></term>
      <listitem><para>enable verbose output</para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-V</option></term>
      <term><option>--version</option></term>
      <listitem><para>display version information, then exit</para></listitem>
     </varlistentry>

     <varlistentry>
      <term><option>-?</option></term>
      <term><option>-h</option></term>
      <term><option>--help</option></term>
      <listitem><para>show help, then exit</para></listitem>
     </varlistentry>

    </variablelist>
   </para>

 </sect2>

 <sect2>
  <title>Upgrade Steps</title>

  <procedure>
   <step performance="optional">
    <title>Optionally move the old cluster</title>

    <para>
     If you are using a version-specific installation directory, e.g.
     <filename>/opt/PostgreSQL/8.4</>, you do not need to move the old cluster. The
     one-click installers all use version-specific installation directories.
    </para>

    <para>
     If your installation directory is not version-specific, e.g.
     <filename>/usr/local/pgsql</>, it is necessary to move the current PostgreSQL install
     directory so it does not interfere with the new <productname>PostgreSQL</> installation.
     Once the current <productname>PostgreSQL</> server is shut down, it is safe to rename the
     PostgreSQL installation directory; assuming the old directory is
     <filename>/usr/local/pgsql</>, you can do:

<programlisting>
mv /usr/local/pgsql /usr/local/pgsql.old
</programlisting>
     to rename the directory.
    </para>
   </step>

   <step>
    <title>For source installs, build the new version</title>

    <para>
     Build the new PostgreSQL source with <command>configure</> flags that are compatible
     with the old cluster. <application>pg_upgrade</> will check <command>pg_controldata</> to make
     sure all settings are compatible before starting the upgrade.
    </para>
   </step>

   <step>
    <title>Install the new PostgreSQL binaries</title>

    <para>
     Install the new server's binaries and support files. You can use the
     same port numbers for both clusters, typically 5432, because the old and
     new clusters will not be running at the same time.
    </para>

    <para>
     For source installs, if you wish to install the new server in a custom
     location, use the <literal>prefix</literal> variable:

<programlisting>
gmake prefix=/usr/local/pgsql.new install
</programlisting>
    </para>
   </step>

   <step>
    <title>Install pg_upgrade and pg_upgrade_support</title>

    <para>
     Install the <application>pg_upgrade</> binary and
     <application>pg_upgrade_support</> library in the new PostgreSQL cluster.
    </para>
   </step>

   <step>
    <title>Initialize the new PostgreSQL cluster</title>

    <para>
     Initialize the new cluster using <command>initdb</command>.
     Again, use compatible <command>initdb</command>
     flags that match the old cluster. Many
     prebuilt installers do this step automatically. There is no need to
     start the new cluster.
    </para>
   </step>

   <step>
    <title>Install custom shared object files</title>

    <para>
     Install any custom shared object files (or DLLs) used by the old cluster
     into the new cluster, e.g. <filename>pgcrypto.so</filename>, whether they are from <filename>contrib</filename>
     or some other source. Do not install the schema definitions, e.g.
     <filename>pgcrypto.sql</>, because these will be upgraded from the old cluster.
    </para>
   </step>

   <step>
    <title>Adjust authentication</title>

    <para>
     <command>pg_upgrade</> will connect to the old and new servers several times,
     so you might want to set authentication to <literal>trust</> in
     <filename>pg_hba.conf</>, or if using <literal>md5</> authentication,
     use a <filename>~/.pgpass</> file (see <xref linkend="libpq-pgpass">)
     to avoid being prompted repeatedly for a password.  Also make sure
     pg_upgrade is the only program that can connect to the clusters.
    </para>
   </step>

   <step>
    <title>Stop both servers</title>

    <para>
     Make sure both database servers are stopped using, on Unix, e.g.:

<programlisting>
pg_ctl -D /opt/PostgreSQL/8.4 stop
pg_ctl -D /opt/PostgreSQL/9.0 stop
</programlisting>

     or on Windows, using the proper service names:

<programlisting>
NET STOP postgresql-8.4
NET STOP postgresql-9.0
</programlisting>

     or

<programlisting>
NET STOP pgsql-8.3  (<productname>PostgreSQL</> 8.3 and older used a different service name)
</programlisting>
    </para>
   </step>

   <step>
    <title>Run <application>pg_upgrade</></title>

    <para>
     Always run the <application>pg_upgrade</> binary of the new server, not the old one.
     <application>pg_upgrade</> requires the specification of the old and new cluster's
     data and executable (<filename>bin</>) directories. You can also specify
     user and port values, and whether you want the data linked instead of
     copied (the default). 
    </para>

    <para>
     If you use link mode, the upgrade will be much faster (no file
     copying), but you will not be able to access your old cluster
     once you start the new cluster after the upgrade.  Link mode also
     requires that the old and new cluster data directories be in the
     same file system.  See <literal>pg_upgrade --help</> for a full
     list of options.
    </para>

    <para>
     For Windows users, you must be logged into an administrative account, and
     then start a shell as the <literal>postgres</> user and set the proper path:

<programlisting>
RUNAS /USER:postgres "CMD.EXE"
SET PATH=%PATH%;C:\Program Files\PostgreSQL\9.0\bin;
</programlisting>

     and then run <application>pg_upgrade</> with quoted directories, e.g.:

<programlisting>
pg_upgrade.exe
        --old-datadir "C:/Program Files/PostgreSQL/8.4/data"
        --new-datadir "C:/Program Files/PostgreSQL/9.0/data"
        --old-bindir "C:/Program Files/PostgreSQL/8.4/bin"
        --new-bindir "C:/Program Files/PostgreSQL/9.0/bin"
</programlisting>

     Once started, <command>pg_upgrade</> will verify the two clusters are compatible
     and then do the upgrade. You can use <command>pg_upgrade --check</>
     to perform only the checks, even if the old server is still
     running. <command>pg_upgrade --check</> will also outline any
     manual adjustments you will need to make after the upgrade.
     <command>pg_upgrade</> requires write permission in the current directory.
    </para>

    <para>
     Obviously, no one should be accessing the clusters during the upgrade.
    </para>

    <para>
     If an error occurs while restoring the database schema, <command>pg_upgrade</> will
     exit and you will have to revert to the old cluster as outlined in <xref linkend="pgupgrade-step-revert">
     below. To try <command>pg_upgrade</command> again, you will need to modify the old
     cluster so the pg_upgrade schema restore succeeds. If the problem is a
     contrib module, you might need to uninstall the contrib module from
     the old cluster and install it in the new cluster after the upgrade,
     assuming the module is not being used to store user data.
    </para>
   </step>

   <step>
    <title>Restore <filename>pg_hba.conf</></title>

    <para>
     If you modified <filename>pg_hba.conf</> to use <literal>trust</>,
     restore its original authentication settings.
    </para>
   </step>

   <step>
    <title>Post-Upgrade processing</title>

    <para>
     If any post-upgrade processing is required, pg_upgrade will issue
     warnings as it completes. It will also generate script files that must
     be run by the administrator. The script files will connect to each
     database that needs post-upgrade processing. Each script should be
     run using:

<programlisting>
psql --username postgres --file script.sql postgres
</programlisting>

     The scripts can be run in any order and can be deleted once they have
     been run.
    </para>

    <caution>
    <para>
     In general it is unsafe to access tables referenced in rebuild scripts
     until the rebuild scripts have run to completion; doing so could yield
     incorrect results or poor performance. Tables not referenced in rebuild
     scripts can be accessed immediately.
    </para>
    </caution>
   </step>

   <step>
    <title>Statistics</title>

    <para>
     Because optimizer statistics are not transferred by <command>pg_upgrade</>, you will
     be instructed to run a command to regenerate that information at the end
     of the upgrade.
    </para>
   </step>

   <step>
    <title>Delete old cluster</title>

    <para>
     Once you are satisfied with the upgrade, you can delete the old
     cluster's data directories by running the script mentioned when
     <command>pg_upgrade</command> completes. You can also delete the
     old installation directories
     (e.g. <filename>bin</>, <filename>share</>).
    </para>
   </step>

   <step id="pgupgrade-step-revert" performance="optional">
    <title>Reverting to old cluster</title>

    <para>
     If, after running <command>pg_upgrade</command>, you wish to revert to the old cluster,
     there are several options:

     <itemizedlist>
      <listitem>
       <para>
        If you ran <command>pg_upgrade</command>
        with <option>--check</>, no modifications were made to the old
        cluster and you can re-use it anytime.
       </para>
      </listitem>

      <listitem>
       <para>
        If you ran <command>pg_upgrade</command>
        with <option>--link</>, the data files are shared between the
        old and new cluster. If you started the new cluster, the new
        server has written to those shared files and it is unsafe to
        use the old cluster.
       </para>
      </listitem>

      <listitem>
       <para>
        If you
        ran <command>pg_upgrade</command> <emphasis>without</> <option>--link</>
        or did not start the new server, the old cluster was not
        modified except that an <literal>.old</> suffix was appended
        to <filename>$PGDATA/global/pg_control</> and perhaps
        tablespace directories. To reuse the old cluster, remove
        the <filename>.old</> suffix
        from <filename>$PGDATA/global/pg_control</>. and, if upgrading
        to 8.4 or earlier, remove the tablespace directories created
        by the upgrade and remove the <filename>.old</> suffix from
        the tablespace directory names; then you can restart the old
        cluster.
       </para>
      </listitem>
     </itemizedlist>
    </para>
   </step>
  </procedure>

 </sect2>

 <sect2>
  <title>Limitations in Upgrading <emphasis>from</> PostgreSQL 8.3</title>

  <para>
   Upgrading from PostgreSQL 8.3 has additional restrictions not present
   when upgrading from later PostgreSQL releases.  For example,
   pg_upgrade will not work for upgrading from 8.3 if a user column
   is defined as:
   <itemizedlist>
    <listitem>
     <para>
      a <type>tsquery</> data type
     </para>
    </listitem>
    <listitem>
     <para>
      data type <type>name</> and is not the first column
     </para>
    </listitem>
   </itemizedlist>
  </para>

  <para>
   You must drop any such columns and upgrade them manually.
  </para>

  <para>
   pg_upgrade will require a table rebuild if:
   <itemizedlist>
    <listitem>
     <para>
      a user column is of data type <type>tsvector</type>
     </para>
    </listitem>
   </itemizedlist>
  </para>

  <para>
   pg_upgrade will require a reindex if:
   <itemizedlist>
    <listitem>
     <para>
      an index is of type hash or GIN
     </para>
    </listitem>
    <listitem>
     <para>
      an index uses <function>bpchar_pattern_ops</>
     </para>
    </listitem>
   </itemizedlist>
  </para>

  <para>
   Also, the default datetime storage format changed to integer after
   <productname>PostgreSQL</> 8.3. pg_upgrade will check that the datetime storage format
   used by the old and new clusters match. Make sure your new cluster is
   built with the configure flag <option>--disable-integer-datetimes</>.
  </para>

  <para>
   For Windows users, note that due to different integer datetimes settings
   used by the one-click installer and the MSI installer, it is only
   possible to upgrade from version 8.3 of the one-click distribution to
   version 8.4 or later of the one-click distribution. It is not
   possible to upgrade from the MSI installer to the one-click installer.
  </para>

 </sect2>

 <sect2>
  <title>Notes</title>

  <para>
   <application>pg_upgrade</> does not support upgrading of databases
   containing these <type>reg*</> OID-referencing system data types:
   <type>regproc</>, <type>regprocedure</>, <type>regoper</>,
   <type>regoperator</>, <type>regclass</>, <type>regconfig</>, and
   <type>regdictionary</>.  (<type>regtype</> can be upgraded.)
  </para>

  <para>
   All failure, rebuild, and reindex cases will be reported by
   <application>pg_upgrade</> if they affect your installation;
   post-upgrade scripts to rebuild tables and indexes will be
   generated automatically.
  </para>

  <para>
   For deployment testing, create a schema-only copy of the old cluster,
   insert dummy data, and upgrade that.
  </para>

  <para>
   If you want to use link mode and you don't want your old cluster
   to be modified when the new cluster is started, make a copy of the
   old cluster and upgrade that with link mode. To make a valid copy
   of the old cluster, use <command>rsync</> to create a dirty
   copy of the old cluster while the server is running, then shut down
   the old server and run <command>rsync</> again to update the copy with any
   changes to make it consistent.
  </para>

 </sect2>

</sect1>