Region Connectors

LakeSentry uses a per-region connector model because Databricks system tables are regional. A collector in East US cannot query system table data for workspaces in West Europe. You need a separate region connector — and a separate collector — for each region where you operate Databricks workspaces.

For initial account setup including your first region, see Account & Connector Setup.

How regions work

Each account connector can have multiple region connectors:

Databricks Account (e.g., Acme Corp)
├── Region: East US
│   └── Collector → reads system tables for East US workspaces
├── Region: West Europe
│   └── Collector → reads system tables for West Europe workspaces
└── Region: West US 2
    └── (not configured yet — no data collected)

LakeSentry automatically aggregates data across all regions into a unified cost view. Workspaces, clusters, jobs, and queries from every region appear together in dashboards and reports.

Why per-region collectors?

Databricks system tables like system.compute.clusters and system.query.history contain data scoped to the region where they are queried. A SQL warehouse running in East US only sees compute and query data for East US workspaces. This is a Databricks platform constraint, not a LakeSentry limitation.

Account-level tables (system.billing.usage, system.billing.list_prices) are an exception — billing data is global and accessible from any region. LakeSentry designates one connector as the primary for global tables, and only that connector ingests billing and other global data to avoid duplicates.

Adding a region

1. Go to Settings > Connector.
2. Click Add Connector.
3. Select the region from the dropdown (e.g., eastus, westeurope, us-west-2).
4. Enter a workspace URL from that region. This workspace is used for validation and as the default collector host.
5. Click Save.

The region connector is created in Pending Setup status until a collector is deployed and reports data.

Workspace URL

The workspace URL identifies which region this connector covers. Use any workspace URL from the target region — LakeSentry validates that the workspace belongs to the configured account and region.

Examples:

• Azure: https://adb-1234567890123456.7.azuredatabricks.net
• AWS: https://dbc-a1b2c3d4-5678.cloud.databricks.com
• GCP: https://1234567890123456.7.gcp.databricks.com

Connection strings

Each region connector has a unique connection string that the collector uses to authenticate and push data to LakeSentry.

Generating a connection string

1. On the Connectors page, find the region connector.
2. Click Generate Connection String.
3. Copy the string (starts with LAKESENTRY://).
4. Store it securely.

The connection string encodes:

Component	Purpose
API URL	Where the collector sends data
Connector ID	Identifies which connector this collector serves
Collector token	One-time authentication token (hashed server-side)
Reference catalog and schema	Catalog and schema for reference data
Mode	Connection mode (e.g., databricks)

Caution

Connection strings contain a one-time token. It is only displayed once at generation time. If you lose it, generate a new one — this invalidates the previous token.

Rotating connection strings

To rotate a connection string (e.g., if you suspect it was compromised):

1. Generate a new connection string on the region connector.
2. Update the collector configuration in Databricks with the new string.
3. The old token is automatically invalidated.

The collector will fail authentication on its next run until you update the connection string. Plan a brief maintenance window if you want to avoid a gap in data collection.

Monitoring region health

The Connectors page shows the health status of each region connector:

Status	Meaning
Active	Connector is running and data was received recently
Pending Setup	Connector is configured but no data has been received yet
Pending Validation	Credentials were updated and are being validated
Inactive	Connector has been disabled
Error	Validation error or persistent failure

Health details

Click on a region connector to see:

• Last ingestion — Timestamp of the most recent data push
• Tables received — Which system tables were successfully extracted
• Collector runs — History of recent collector executions with status and duration
• Extraction checkpoints — Current watermark positions for each table

Managing regions

Removing a region

1. Stop the collector — Disable or delete the Databricks job in that region.
2. On the Connector settings tab, click the region connector’s menu and select Remove.
3. Confirm the deletion.

Caution

Deleting a region connector removes raw ingestion data and connector metadata for that region. Transformed ledger data (dashboards, reports) is preserved. If you only want to pause collection, stop the Databricks job instead.

Re-adding a region

If you previously removed a region and want to re-add it:

1. Add the region connector again (same steps as adding a new region).
2. Generate a new connection string.
3. Deploy or reconfigure the collector.

The collector starts fresh — it does not recover data from before the region was removed.

Multi-cloud considerations

If your Databricks account spans multiple cloud providers (e.g., Azure for production, AWS for development), each cloud region needs its own region connector. The setup process is identical — the cloud provider is determined by the workspace URL.

LakeSentry normalizes data across cloud providers. Cost metrics, attribution rules, and insights work uniformly regardless of which cloud hosts the underlying workspace.

Next steps

• Collector Deployment — Deploy the collector in each region
• Collector Troubleshooting — Diagnosing region-specific issues
• Account & Connector Setup — Initial account connection setup