From 72443572bf5dd147f50b0168e1078d88476a3e9f Mon Sep 17 00:00:00 2001
From: Erik Johnston <erik@matrix.org>
Date: Wed, 29 Apr 2015 11:50:18 +0100
Subject: [PATCH] Mention that postgres databases must have the correct charset
 encoding

---
 docs/postgres.rst | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/docs/postgres.rst b/docs/postgres.rst
index a04ab12b1..d645e1d69 100644
--- a/docs/postgres.rst
+++ b/docs/postgres.rst
@@ -1,6 +1,23 @@
 Using Postgres
 --------------
 
+Set up database
+===============
+
+The PostgreSQL database used *must* have the correct encoding set, otherwise
+would not be able to store UTF8 strings. To create a database with the correct
+encoding use, e.g.::
+
+ CREATE DATABASE synapse
+  ENCODING 'UTF8'
+  LC_COLLATE='C'
+  LC_CTYPE='C'
+  template=template0
+  OWNER synapse_user;
+
+This would create an appropriate database named ``synapse`` owned by the
+``synapse_user`` user (which must already exist).
+
 Set up client
 =============