Database Connections

Database Connections

Once you add a database connection, it will be made available from the Add Data Model workflow and you'll be able to specify tables or SQL queries to start visualizing data in Hashboard.

Firewall Configuration

If you use a firewall to restrict access to your database, you'll need to allowlist Hashboard's IPs.
Hashboard will always connect to your database from one of these IP addresses.

SSH tunneling

Currently in beta, Hashboard supports connecting to a Postgres or Redshift database via a bastion host. A bastion host is a server which you control that has access to your database but is not the database host itself. This may be useful when handling especially sensitive data. To join the beta, [contact us]( tunnel beta).

To use a bastion host, toggle on Use SSH tunnel when creating a Postgres database connection. Fill out the additional credentials according to your bastion host set up. Download Hashboard's public ssh key and set it as an authorized key on your server under the bastion host user that you specify in the data connection.

Fill out the remaining credentials as they'll be accessed by your bastion host, e.g. for Port specify the port on which your bastion host connects to the database.

Create a database connection

There are two ways to create a database connection in Hashboard: using the web interface or the CLI. Follow these steps to create a database connection through the UI:

  1. Go to the Data sources (opens in a new tab) page using the link in the project dropdown
  2. Click + Add connection
  3. Select the type of database
  4. Fill out the credentials according to your type of database
  5. Click the antenna icon to test your connection 🗼
  6. Click Add to complete the process

Specifying Schemas For most database types, you can optionally specify a schema. If you specify a schema in Hashboard, only tables from that schema will be made available in the Add Data Model workflow

As an alternative, you can create a database connection using the CLI using the hb datasource create command. For more information, see the CLI documentation.