Guide: Using PostgreSQL with Databox
PostgreSQL connections are available in Plus and Business Accounts. Agency Basic Accounts have the ability to use PostgreSQL in the Agency Account only.
How to prepare your PostgreSQL database
IN THIS SECTION
How to prepare your PostgreSQL database
To prepare our PostgreSQL database and host, we'll need to allow remote connections from Databox’s public IP (184.108.40.206) to our Database. The necessary steps to do that will depend on your database, server infrastructure, and firewall. Our IP must be able to connect to the database port directly.
For our purposes, we'll assume our PostgreSQL server is up and running, reachable from internet addresses and accepting TCP/IP connections on some TCP port (5432 is the default port).
CREATE USER databox_ro WITH PASSWORD 'secretPassword'; GRANT ALL ON DATABASE mydb to databox_ro; GRANT SELECT ON ALL TABLES IN SCHEMA public TO databox_ro; -- at least 9.0 GRANT SELECT ON ALL SEQUENCES IN SCHEMA public TO databox_ro; -- at least 9.0<br>
To keep this simple, we have given this user SELECT permission for all Tables in our database. Permission could be given to select from one Table only, or we could create a custom view and give permissions to select from this view only. You can create a Custom View and give the user permission to select only select from this View only. A basic example of how to achieve this can be found here.
/etc/postgresql/9.5/main/postgresql.conf(default for Ubuntu and Debain). Check to ensure it contains the following:
listen_addresses = '*' # Listen on all host IP addresses available port = 5432 # Listen on this port, default is 5432<br>
service postgresql restart
pg_hba.confand reload the database:
host mydb databox_ro 220.127.116.11 255.255.255.255 md5
iptables -A FORWARD -s 18.104.22.168/32 -p tcp --dport 5432 -j ACCEPT<br>
You may have to replace FORWARD with INPUT if your Linux box has a public IP and the Database runs on the Host itself.
Our server is now set up to accept requests from Databox's IP to our Database/ Table.
How to set up your SSL connection
It’s a good practice to secure the connection with an SSL certificate that is generated and installed on the server.
The PostgreSQL server can be configured with SSL enabled by setting the parameter SSL to on in
postgresql.conf. The server will listen for both normal and SSL connections on the same TCP port, and will negotiate with any connecting client on whether to use SSL.
How to connect PostgreSQL to Databox
Once our Database is properly prepared, we'll connect it to Databox.
To connect PostgreSQL to Databox, go to Data Sources > + New connection. Type PostgreSQL in the search bar and click on the green Connect button.
This will open the Activation window, which is where you will be prompted to enter your connection data and authorize the Databox app.
If you have set up a SSL certificate, be sure to check the SSL checkbox and paste the certificates in the corresponding fields. SSL CA is not mandatory. If you're using a self-signed certificate, be sure to leave the SSL checkbox un-checked or the connection will fail.
Click the green Activate button to complete the connection process. If PostgreSQL was successfully connected, the Activation window will close shortly and you'll see a Connected notice.
If you see a notice that says Wrong Credentials, doublecheck your user data before re-entering them in the Activation window.
If the Activation window doesn't close for over a minute, there may be difficulty connecting to your Database Host due to firewall, server, or networking issues.
If you have difficulty connecting, be sure the IP or hostname entered is publicly available or access is granted to Databox's public IP (22.214.171.124).
How to use the Query Builder for PostgreSQL
The Query Builder for PostgreSQL allows you to create Custom Metrics using data from your Database to use in Databoards, Alerts, Goals, etc. Learn more about the Query Builder for PostgreSQL here.
- Query results are limited to 1000 rows. If you expect your query to return over 1000 rows or if you're not sure how many rows will be returned, be sure to LIMIT the query
- The query must contain a column named "Date." If you need to, use AS in your SELECT statement to satisfy this requirement