Connect to Databricks to query and analyze your lakehouse data. Requires a SQL warehouse and personal access token.
Prerequisites
You need a running SQL warehouse in your Databricks workspace. Starting a warehouse can take 2-5 minutes.
Connect to Databricks
- Click Add data in the Layers panel
- Go to the Connections tab
- Select Databricks
- Fill in the connection parameters
- Click Create Connection
Connection Parameters
| Parameter | Required | Description |
|---|---|---|
| Connection Name | Yes | A friendly name to identify this connection |
| Server Hostname | Yes | Workspace hostname from your Databricks URL |
| HTTP Path | Yes | Path to your SQL warehouse |
| Access Token | Yes | Personal access token or service principal token |
| Catalog | No | Unity Catalog name (recommended for performance) |
| Default Schema | No | Default schema within the catalog |
Finding Connection Details
- Open your Databricks workspace
- Go to SQL Warehouses and select your warehouse
- Click Connection details
- Copy the Server hostname and HTTP path
Server hostname format: dbc-a1b2c3d4-e5f6.cloud.databricks.com
HTTP path format: /sql/1.0/warehouses/abc123def456
Generating an Access Token
- Click your username in the top-right corner
- Go to User Settings → Developer → Access Tokens
- Click Generate new token
- Set an expiration and click Generate
- Copy the token (starts with
dapi)
Working with Data
After connecting, browse available catalogs, schemas, and tables. Select the data you want to import and click Add to Project. Delta tables with geometry columns import as vector layers.
IP Access Lists
If your Databricks workspace uses IP access lists, add these Atlas IP addresses:
16.16.129.130
51.20.23.178
13.50.71.210
51.21.188.73
Troubleshooting
| Issue | Solution |
|---|---|
| Connection Timeout | Warehouse may be starting. Wait 2-5 minutes and retry. |
| Cluster Unavailable | Start the SQL warehouse in Databricks and wait for it to be ready. |
| Authentication Failed | Verify access token is valid and not expired. |
| Endpoint Not Found | Check server hostname and HTTP path are correct. |
| Cluster Terminated / Stopped | Restart the warehouse in Databricks workspace. |