This article explains how to install and configure PostgreSQL for a Confluence on a CentOS 7 server. Read the other Confluence articles on this blog to ensure your web server (if you’re using one) is configured correctly.
yum install postgresql-server postgresql-contrib
Next you’ll need to do the automatic setup:
Now set the permissions for the database server:
host all all 127.0.0.1/32 ident host all all ::1/128 ident
With this (not very secure but is a quick start):
host all all 127.0.0.1/32 trust host all all ::1/128 trust
Now enable and start the PostgreSQL server:
systemctl enable postgresql systemctl start postgresql
Now we create the database ready for Confluence. Switch to the appropriate user and log in:
sudo -i -u postgres
Create the new database with the “utf-8” encoding as is required by Confluence. Mine is called “confluencedb”. This command is executed from Bash as the “postgresql” user.
createdb confluencedb --encoding='utf-8' --locale=en_US.utf8 --template=template0
And create a new user called “confluence”. The following will allow you to interactively create a new user. This command is executed from Bash as the “postgresql” user.
Grant permissions to your new user to the database. First start the “psql” CLI tool.
And then execute this command to grant all permissions:
GRANT ALL PRIVILEGES ON DATABASE confluencedb to confluence;
At this point we’re done. You can point Confluence at this new PostgreSQL database and expect it to work.
Within the Confluence installation process, the question of which database to connect to (in Production) should be “PostgreSQL” and then set the “By Connection String” field to the following. Make sure to change the database name form “confluencedb” to whatever you created earlier:
Psql command tips
List all users (from within psql):
List all databases (from within psql):
Create a new user and password (from bash as postgres):
createuser -P myuser
Delete a user (from bash as postgres):