synchronous_standby_names

doc html

<p>
        Specifies a list of standby servers that can support
        <em class="firstterm">synchronous replication</em>, as described in
        <a class="xref" href="/docs/18/warm-standby.html#SYNCHRONOUS-REPLICATION" title="26.2.8. Synchronous Replication">Section 26.2.8</a>.
        There will be one or more active synchronous standbys;
        transactions waiting for commit will be allowed to proceed after
        these standby servers confirm receipt of their data.
        The synchronous standbys will be those whose names appear
        in this list, and
        that are both currently connected and streaming data in real-time
        (as shown by a state of <code class="literal">streaming</code> in the
        <a class="link" href="/docs/18/monitoring-stats.html#MONITORING-PG-STAT-REPLICATION-VIEW" title="27.2.4. pg_stat_replication">
<code class="structname">pg_stat_replication</code></a> view).
        Specifying more than one synchronous standby can allow for very high
        availability and protection against data loss.
       </p><p>
        The name of a standby server for this purpose is the
        <code class="varname">application_name</code> setting of the standby, as set in the
        standby's connection information.  In case of a physical replication
        standby, this should be set in the <code class="varname">primary_conninfo</code>
        setting; the default is the setting of <a class="xref" href="/docs/18/runtime-config-logging.html#GUC-CLUSTER-NAME">cluster_name</a>
        if set, else <code class="literal">walreceiver</code>.
        For logical replication, this can be set in the connection
        information of the subscription, and it defaults to the
        subscription name.  For other replication stream consumers,
        consult their documentation.
       </p><p>
        This parameter specifies a list of standby servers using
        either of the following syntaxes:
</p><pre class="synopsis">
[FIRST] <em class="replaceable"><code>num_sync</code></em> ( <em class="replaceable"><code>standby_name</code></em> [, ...] )
ANY <em class="replaceable"><code>num_sync</code></em> ( <em class="replaceable"><code>standby_name</code></em> [, ...] )
<em class="replaceable"><code>standby_name</code></em> [, ...]
</pre><p>
        where <em class="replaceable"><code>num_sync</code></em> is
        the number of synchronous standbys that transactions need to
        wait for replies from,
        and <em class="replaceable"><code>standby_name</code></em>
        is the name of a standby server.
        <em class="replaceable"><code>num_sync</code></em>
        must be an integer value greater than zero.
        <code class="literal">FIRST</code> and <code class="literal">ANY</code> specify the method to choose
        synchronous standbys from the listed servers.
       </p><p>
        The keyword <code class="literal">FIRST</code>, coupled with
        <em class="replaceable"><code>num_sync</code></em>, specifies a
        priority-based synchronous replication and makes transaction commits
        wait until their WAL records are replicated to
        <em class="replaceable"><code>num_sync</code></em> synchronous
        standbys chosen based on their priorities. For example, a setting of
        <code class="literal">FIRST 3 (s1, s2, s3, s4)</code> will cause each commit to wait for
        replies from three higher-priority standbys chosen from standby servers
        <code class="literal">s1</code>, <code class="literal">s2</code>, <code class="literal">s3</code> and <code class="literal">s4</code>.
        The standbys whose names appear earlier in the list are given higher
        priority and will be considered as synchronous. Other standby servers
        appearing later in this list represent potential synchronous standbys.
        If any of the current synchronous standbys disconnects for whatever
        reason, it will be replaced immediately with the next-highest-priority
        standby. The keyword <code class="literal">FIRST</code> is optional.
       </p><p>
        The keyword <code class="literal">ANY</code>, coupled with
        <em class="replaceable"><code>num_sync</code></em>, specifies a
        quorum-based synchronous replication and makes transaction commits
        wait until their WAL records are replicated to <span class="emphasis"><em>at least</em></span>
<em class="replaceable"><code>num_sync</code></em> listed standbys.
        For example, a setting of <code class="literal">ANY 3 (s1, s2, s3, s4)</code> will cause
        each commit to proceed as soon as at least any three standbys of
        <code class="literal">s1</code>, <code class="literal">s2</code>, <code class="literal">s3</code> and <code class="literal">s4</code>
        reply.
       </p><p>
<code class="literal">FIRST</code> and <code class="literal">ANY</code> are case-insensitive. If these
        keywords are used as the name of a standby server,
        its <em class="replaceable"><code>standby_name</code></em> must
        be double-quoted.
       </p><p>
        The third syntax was used before <span class="productname">PostgreSQL</span>
        version 9.6 and is still supported. It's the same as the first syntax
        with <code class="literal">FIRST</code> and
        <em class="replaceable"><code>num_sync</code></em> equal to 1.
        For example, <code class="literal">FIRST 1 (s1, s2)</code> and <code class="literal">s1, s2</code> have
        the same meaning: either <code class="literal">s1</code> or <code class="literal">s2</code> is chosen
        as a synchronous standby.
       </p><p>
        The special entry <code class="literal">*</code> matches any standby name.
       </p><p>
        There is no mechanism to enforce uniqueness of standby names.  In case
        of duplicates one of the matching standbys will be considered as
        higher priority, though exactly which one is indeterminate.
       </p><div class="note"><h3 class="title">Note</h3><p>
         Each <em class="replaceable"><code>standby_name</code></em>
         should have the form of a valid SQL identifier, unless it
         is <code class="literal">*</code>.  You can use double-quoting if necessary.  But note
         that <em class="replaceable"><code>standby_name</code></em>s are
         compared to standby application names case-insensitively, whether
         double-quoted or not.
        </p></div><p>
        If no synchronous standby names are specified here, then synchronous
        replication is not enabled and transaction commits will not wait for
        replication.  This is the default configuration.  Even when
        synchronous replication is enabled, individual transactions can be
        configured not to wait for replication by setting the
        <a class="xref" href="/docs/18/runtime-config-wal.html#GUC-SYNCHRONOUS-COMMIT">synchronous_commit</a> parameter to
        <code class="literal">local</code> or <code class="literal">off</code>.
       </p><p>
        This parameter can only be set in the <code class="filename">postgresql.conf</code>
        file or on the server command line.
       </p>

documentation revision

{"inputs":[{"file":"runtime-config-replication.html","kind":"same-version-reviewed-manual","locale":"en","sha256":"c0b8bd86131ec62d4df97ce480d004aa7e3544477d053f0c2d4047b9edad5cc0","version":"18.7"},{"file":"runtime-config-replication.html","kind":"same-version-reviewed-manual","locale":"zh-Hans","sha256":"83907b5f469e979a7b9ec875b5e7bdaa2ef6b85596ca31b45174f68cd7398b18","version":"18.7"}],"release":"18.7","runtime_remeasured":false,"source_revision":"pgdoc:b9f0e7232f60f380d642ac9b99226606c254b552"}