<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>31.5. Architecture</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="logical-replication-restrictions.html" title="31.4. Restrictions" /><link rel="next" href="logical-replication-monitoring.html" title="31.6. Monitoring" /></head><body><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">31.5. Architecture</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="logical-replication-restrictions.html" title="31.4. Restrictions">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="logical-replication.html" title="Chapter 31. Logical Replication">Up</a></td><th width="60%" align="center">Chapter 31. Logical Replication</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 11.5 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="logical-replication-monitoring.html" title="31.6. Monitoring">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="LOGICAL-REPLICATION-ARCHITECTURE"><div class="titlepage"><div><div><h2 class="title" style="clear: both">31.5. Architecture</h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="logical-replication-architecture.html#LOGICAL-REPLICATION-SNAPSHOT">31.5.1. Initial Snapshot</a></span></dt></dl></div><p>
Logical replication starts by copying a snapshot of the data on the
publisher database. Once that is done, changes on the publisher are sent
to the subscriber as they occur in real time. The subscriber applies data
in the order in which commits were made on the publisher so that
transactional consistency is guaranteed for the publications within any
single subscription.
</p><p>
Logical replication is built with an architecture similar to physical
streaming replication (see <a class="xref" href="warm-standby.html#STREAMING-REPLICATION" title="26.2.5. Streaming Replication">Section 26.2.5</a>). It is
implemented by <span class="quote">“<span class="quote">walsender</span>”</span> and <span class="quote">“<span class="quote">apply</span>”</span>
processes. The walsender process starts logical decoding (described
in <a class="xref" href="logicaldecoding.html" title="Chapter 49. Logical Decoding">Chapter 49</a>) of the WAL and loads the standard
logical decoding plugin (pgoutput). The plugin transforms the changes read
from WAL to the logical replication protocol
(see <a class="xref" href="protocol-logical-replication.html" title="53.5. Logical Streaming Replication Protocol">Section 53.5</a>) and filters the data
according to the publication specification. The data is then continuously
transferred using the streaming replication protocol to the apply worker,
which maps the data to local tables and applies the individual changes as
they are received, in correct transactional order.
</p><p>
The apply process on the subscriber database always runs with
<code class="varname">session_replication_role</code> set
to <code class="literal">replica</code>, which produces the usual effects on triggers
and constraints.
</p><p>
The logical replication apply process currently only fires row triggers,
not statement triggers. The initial table synchronization, however, is
implemented like a <code class="command">COPY</code> command and thus fires both row
and statement triggers for <code class="command">INSERT</code>.
</p><div class="sect2" id="LOGICAL-REPLICATION-SNAPSHOT"><div class="titlepage"><div><div><h3 class="title">31.5.1. Initial Snapshot</h3></div></div></div><p>
The initial data in existing subscribed tables are snapshotted and
copied in a parallel instance of a special kind of apply process.
This process will create its own temporary replication slot and
copy the existing data. Once existing data is copied, the worker
enters synchronization mode, which ensures that the table is brought
up to a synchronized state with the main apply process by streaming
any changes that happened during the initial data copy using standard
logical replication. Once the synchronization is done, the control
of the replication of the table is given back to the main apply
process where the replication continues as normal.
</p></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="logical-replication-restrictions.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="logical-replication.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="logical-replication-monitoring.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">31.4. Restrictions </td><td width="20%" align="center"><a accesskey="h" href="index.html">Home</a></td><td width="40%" align="right" valign="top"> 31.6. Monitoring</td></tr></table></div></body></html>
