Custom JDBC Configs - Databricks Example

Overview

Starting from QueryPie ACP 11.6.0, the Custom JDBC Configs feature allows administrators to register a JDBC Driver and create a Databricks connection with policies applied.
The query parser (QSI) supported by QueryPie ACP for Databricks is ANSI query, so Databricks-specific syntax may not work correctly.

Registering Databricks as a Custom JDBC Vendor

1. Navigate to Admin > Databases > Connection Management > Custom JDBC Configs.
2. Click the Create Custom JDBC button in the upper right corner.
3. Enter an identifiable name in the Name field. This name will be displayed in the Custom JDBC Vendors section when creating a DB connection.
4. Enter any additional information in the Description field.
5. Enter the Databricks driver class in the Driver Class field of the JDBC Configuration section. (For Databricks: com.databricks.client.jdbc.Driver)
   In general, the Driver Class can be verified with the following command:

unzip -p <your-driver.jar> META-INF/services/java.sql.Driver

6. Enter jdbc:databricks://{host}:{port}/default in the JDBC URL Template.
7. Enter / in the Options Prefix.
8. Enter 443 in the Default Port.
9. Enter --add-opens=java.base/java.nio=ALL-UNNAMED in the JVM Arguments.
10. Click the Select JAR file to upload button in the Driver JAR File section and upload the JDBC Driver.
    Databricks: JDBC Driver Download 
11. Click the Save button to save the configuration.

Creating a Databricks Connection with a Custom Vendor

1. Navigate to Admin > Databases > Connection Management > DB Connections.
2. Click the Create Connection button in the upper right corner.
3. Scroll down and select the Databricks entry you registered earlier (the item entered as Name when creating the Custom JDBC) from the Custom JDBC Vendors section.
4. Enter the connection information such as Connection Name, Host, Port, etc.
   1. Host: Copy and paste the server hostname value obtained from Databricks.
   2. Port: Use the default value of 443 specified in the Custom Vendor.
   3. Database Name: Leave it empty or enter the name corresponding to a Databricks Catalog (refer to the Structure Mapping Comparison table below). Leaving it empty is recommended since this specifies the default path.
   4. Username: Since Databricks uses tokens, leave the Username empty.
   5. Password: Enter the token value issued from Settings (User Settings) > Developer > Access tokens in the Databricks console. When generating the token, specify the token scope as sql, workspace, catalog.
   6. Other Options: Copy and paste the portion after default; from the JDBC URL field found in the connection details of the Databricks console.
5. Select a QSI Parser if needed. Select the default ANSI SQL.
6. Click the Next button to proceed to the connection creation complete screen.

Structure Mapping Comparison