Skip to content

Connecting a New Data Source

Creating a new data source is available through the interface: Dashboard > Catalog > Data Sources > Add or Dashboard > Create > Utilities > Data Source

new_data_source

Registration parameters:

  • Name (required field) - the name of the data source in the system. It must contain only Latin characters, numbers, and hyphens. A secret will be created in the namespace where Predicate is deployed based on this name, so this name must not match existing secrets.
  • Description (optional field) - a description of the data source.
  • Tags (optional field) - a list of tags for the data source.

After filling in the name, you must click the "Validate" button. If a secret with the specified name already exists, an error will appear. If the name is available, additional fields for secret parameterization will appear.

new_data_source

Secret parameters:

  • Connection Name (required field) - the name of the data source. The following options are available:

    • s3-duckdb - connection to S3 source;
    • postgresql - connection to PostgreSQL DB;
    • oracle - connection to Oracle DB;
    • hive - connection to Hive DB;
    • impala - connection to Impala DB.

  • Driver (required field) - the driver for connecting to the source, automatically filled after selecting the connection name.

  • Classname (required field) - the class for connecting to the source, automatically filled after selecting the connection name.
  • URL (required field) - the connection path to the source. For s3-duckdb, it is filled automatically; for other connections, a data entry template appears.
  • Driver Arguments - a list of driver parameter values. The specific list appears after selecting the connection name.
  • Tags - a list of automatic metrics assigned depending on the connection name.

Depending on the connection name, Driver Arguments may contain the following fields:

Connection Name Driver Arguments
s3-duckdb 1. Path to S3 (s3_url)
2. Connection Key (s3_access_key)
3. Connection Secret (s3_secret_access_key)
4. Bucket Name (s3_bucket)
postgresql 1. Login (user)
2. Password (password)
oracle 1. Login (user)
2. Password (password)
hive 1. Login (user)
2. Password (password)

After filling in all the data, you must click the "Test Connection" button to verify the entered data. After a successful connection, you must click the "Create" button.

Registration result: the data source parameters become available through Dashboard > Catalog > Data Sources. The list of sources displays the connection success status.