🧱 Databricks Connection

Connect your Databricks lakehouse to Meza AI.

Overview

Connect your Databricks workspace to Meza AI to query data from your lakehouse. This integration uses Databricks SQL warehouses for efficient query execution against Delta Lake tables.

Prerequisites

  • Databricks workspace (AWS, Azure, or GCP)
  • SQL warehouse (serverless or pro) enabled
  • Personal access token with SQL permissions
  • Catalog and schema access configured

Creating an Access Token

1

Go to User Settings

In Databricks, click your username in the top-right → User Settings.

2

Open Access Tokens

Go to the DeveloperAccess Tokens tab.

3

Generate Token

Click Generate New Token, enter a comment like "Meza AI", and set an expiration.

4

Copy Token

Copy the token immediately — you won't be able to see it again!

Finding Connection Details

To find your SQL warehouse connection details:

  1. Go to SQL Warehouses in your Databricks workspace
  2. Select your warehouse
  3. Click Connection details tab
  4. Copy the Server Hostname and HTTP Path

Connection Steps

1

Navigate to Databases

Go to ConfigurationDatabases in the left sidebar.

2

Select Databricks

Click Add Connection and select Databricks.

3

Enter Connection Details

Enter server hostname, HTTP path, and your access token.

4

Configure Catalog

Enter your Unity Catalog name and default schema.

5

Test & Save

Click Test Connection then Save.

Connection Parameters

ParameterDescriptionExample
Server HostnameWorkspace URL without https://dbc-abc123.cloud.databricks.com
HTTP PathSQL warehouse endpoint path/sql/1.0/warehouses/abc123def
Access TokenPersonal access tokendapi...
CatalogUnity Catalog namemain
SchemaDefault schemadefault

💡 Note

The HTTP Path can be found in your SQL warehouse's Connection Details tab. It starts with /sql/1.0/warehouses/.

Required Permissions

The access token user needs these permissions:

  • Can Use permission on the SQL warehouse
  • USE CATALOG on the Unity Catalog
  • USE SCHEMA on target schemas
  • SELECT on tables you want to query

Troubleshooting

Connection Timeout

  • Check if the SQL warehouse is running (it may be stopped)
  • Verify the server hostname is correct
  • Ensure network connectivity to Databricks

Authentication Error

  • Verify the access token hasnt expired
  • Check the token has required permissions
  • Generate a new token if needed

What's Next?