Skip to content

Databricks

Overview

Connect Databricks to Trust3 to enable natural language queries across your data lakehouse. Trust3 integrates seamlessly with Databricks SQL warehouses using personal access token authentication.

Prerequisites

Requirement Description
Databricks Workspace Active Databricks workspace
Access Token Personal access token with full read permissions on target database(s)
SQL Warehouse SQL warehouse HTTP path

Quick Setup Process

  1. Navigate to Data Sources from the main menu → Click Add Data Source
  2. Select your data source type from the Source dropdown
  3. Enter a Connection Name (e.g., Sales Database)
  4. Configure the connection details
  5. Click Test Connection to validate
  6. Once successful, click Add Data Source to save

Finding Required Information

Finding Your HTTP Path

To find your SQL warehouse HTTP path:

  1. Log in to your Databricks workspace
  2. Navigate to SQL Warehouses from the sidebar
  3. Click on your SQL warehouse name
  4. Go to the Connection Details tab
  5. Copy the HTTP Path value (format: /sql/1.0/warehouses/<warehouse-id>)

Configuration Details

Basic Information

Field Description Example
Connection Name * A unique identifier for this data source connection Production [Your Data Source]

Connection Settings

Field Description Example
Server Hostname * Your Databricks workspace URL (without https:// prefix) your-workspace.cloud.databricks.com
HTTP Path * Path to your SQL warehouse (find in SQL warehouse configuration) /sql/1.0/warehouses/abc123def456
Access Token * Databricks personal access token (starts with dapi...) dapi1234567890abcdef...

Generating Personal Access Tokens

To create a personal access token:

  1. Log in to Databricks workspace → User Settings → Developer → Access Tokens
  2. Click "Generate New Token"
  3. Set a comment and lifetime (optional)
  4. Click Generate and copy the token immediately

📘 Reference: Databricks Personal Access Token Authentication

Testing the Connection

After entering your connection details click Test Connection to validate the configuration

Connection Test Successful

Once the test is successful, you can proceed to save the data source.

Connection Test Failed

If the test fails, check the Troubleshooting section below for common issues and solutions.

Managing Data Sources

Update Connection

To update your data source connection:

  1. Navigate to Data Sources from the main menu
  2. Locate the connection you want to update
  3. Click the Actions button (three dots) → Edit
  4. Modify the connection fields as needed
  5. Click Test Connection to validate changes
  6. Click Update Data Source to save

Delete Connection

To delete an existing data source connection:

  1. Navigate to Data Sources from the main menu
  2. Locate the connection you want to delete
  3. Click the Actions button (three dots) → Delete
  4. Confirm the deletion in the dialog

Permanent Deletion Warning

Deleting a data source is permanent and cannot be undone. Any queries, reports, or integrations using this data source will stop working immediately. Make sure you want to proceed before confirming the deletion.

Troubleshooting

Common Issues

Issue Common Causes Solution
Invalid hostname Wrong format or https:// prefix included Use format: your-workspace.cloud.databricks.com (no protocol)
Authentication failed Wrong or expired token Generate a new personal access token
HTTP path not found Incorrect warehouse ID or warehouse doesn't exist Verify warehouse ID in HTTP path
Warehouse is stopped SQL warehouse not running Start the SQL warehouse in Databricks
Insufficient privileges Token lacks read permissions Grant necessary read permissions to the user
Network timeout Firewall or network policy blocking connection Check firewall rules and network policies
Token revoked Access token manually revoked Generate a new access token

Additional Checks

  • Ensure hostname uses correct cloud provider domain (.cloud.databricks.com)
  • Verify SQL warehouse exists and is accessible
  • Confirm token starts with dapi and hasn't expired
  • Check user account associated with token is active
  • Verify token has appropriate workspace access permissions
  • Ensure SQL warehouse is in Running state (not stopped or paused)

Next Steps