How to Change Gateway Connection in Power BI

Switching your Power BI data source from one gateway to another is a common task, especially as your data infrastructure evolves. Whether you're moving a database to a new server or centralizing your connections, knowing how to properly remap a dataset is essential for keeping your reports live and accurate. This guide will walk you through the process step-by-step and cover common troubleshooting tips to ensure a smooth transition.

Graphed

Still Building Reports Manually?

Watch how growth teams are getting answers in seconds — not days.

Watch the Demo

When and Why You Might Need to Change a Gateway Connection

Before jumping into the how-to, it’s helpful to understand the scenarios where this change becomes necessary. You're not just doing this for fun, there's always a practical reason behind updating your gateway configuration. Here are the most common situations:

• Server Migration or Upgrade: One of the most frequent reasons is moving an on-premises data source, like a SQL Server or an Oracle database, to a new physical or virtual machine. The old gateway points to the old server, so you need to point your reports to a new gateway installed on or connected to the new server.
• Gateway Consolidation: Your organization might start with team members using several "personal mode" gateways. For security, stability, and easier management, businesses often consolidate these into a few standard "on-premises data gateways" (in a cluster for high availability). This requires migrating datasets from personal gateways to the central one.
• Switching Environments (Dev to Prod): In a professional BI setup, you typically have separate environments for development, testing, and production. Each environment should have its own gateway. When a report is ready to go live, you’ll need to republish it and change its data source connection from the development gateway to the production gateway.
• Performance Troubleshooting or Load Balancing: If an existing gateway is overloaded or experiencing connectivity issues, you might need to shift some datasets to a different gateway. Using a gateway cluster is the best practice here, as it automatically handles load balancing and failover, but sometimes a manual switch is necessary for specific reports.

Before You Begin: Key Prerequisites

Success with this task is all about preparation. Trying to change the gateway without having the new path set up first will only lead to frustration and error messages. Before you even open your Power BI workspace, make sure you have the following in place:

• The New Gateway is Installed and Configured: The gateway you want to switch to must already be installed on a server, registered with your Power BI tenant, and showing as "Online." It's best practice to ensure it's the latest version available from Microsoft.
• The Data Source is Registered on the NEW Gateway: This is the step most people miss. You must go into the Power BI service (under Settings > Manage connections and gateways) and add your data source (e.g., SQL server & database details) to the new gateway. The credentials must be tested and working.
• Admin Permissions: You need to be an admin on the new gateway cluster or, at a minimum, have permissions to use the data source registration on that gateway. You will also need to be an owner or member of the workspace containing the dataset you're updating.
• Matching Credentials: The credentials you use to register the data source on the new gateway must have the necessary permissions to access the target database or file share. If they can't connect, your refresh will fail.

Graphed

Still Building Reports Manually?

Watch how growth teams are getting answers in seconds — not days.

Watch the Demo

Step-by-Step: Changing the Gateway for an Existing Dataset

Once your prerequisites are sorted, the actual process of remapping the connection is surprisingly straightforward. It only takes a few clicks within the Power BI Service settings.

Step 1: Locate Your Dataset in the Power BI Service

First, log in to your Power BI account at app.powerbi.com. In the navigation pane on the left, navigate to the workspace that contains the report and dataset you want to update. Once inside the workspace, switch from the "All" or "Reports" view to focus just on the "Datasets + dataflows" tab. Find the specific dataset that needs its gateway changed. Remember, you're changing the setting on the dataset, not the report itself.

Step 2: Access the Dataset Settings

Hover over the dataset and click the three vertical dots (…) to open the "More options" menu. From this menu, select Settings. This will take you to the control panel for this individual dataset, where you can manage everything from refresh schedules to credentials.

Step 3: Find the "Gateway and cloud connections" Section

On the Settings page, scroll down until you see a section explicitly named Gateway and cloud connections. Power BI will automatically list all the on-premises data sources detected in your dataset’s Power Query connections. You'll see one or more data sources listed here, along with the gateway they are currently mapped to.

Graphed

Still Building Reports Manually?

Watch how growth teams are getting answers in seconds — not days.

Watch the Demo

Step 4: Map the Data Source to the New Gateway

Each data source in this section will have a dropdown menu under the title "Maps to." This dropdown is the key to the entire process. It will show a list of all available and compatible data sources that you've configured across all gateways you have access to.

Click the dropdown. Carefully select the data source that corresponds to the one configured on your newly prepared gateway. Power BI identifies compatible data sources based on connection details like server and database names, so consistency is vital.

Important Note: If you do not see your new data source connection in the list, it's almost always because of one of two reasons:

1. You haven't added the data source directly to the new gateway in the Power BI Service yet (see the Prerequisites section).
2. The connection details (server name, database name, etc.) defined in Power BI Desktop for this report do not perfectly match the details used when registering the source on the new gateway. They must be identical.

After selecting the correct new mapping, click the Apply button.

Step 5: Test and Refresh Your Data

After you apply the change, Power BI will use the new gateway for all future refreshes. As a final validation step, it's wise to trigger a manual data refresh immediately. Navigate back to your workspace view, find the same dataset, and click the circular arrow icon for "Refresh now."

If the refresh completes successfully, congratulations! You've successfully switched your gateway connection. If it fails, head to the troubleshooting section below.

Graphed

Still Building Reports Manually?

Watch how growth teams are getting answers in seconds — not days.

Watch the Demo

Troubleshooting Common Gateway Switching Problems

Encountering a hiccup or two is normal. Here are the most common issues people face and how to resolve them:

• Issue: My new data source doesn't appear in the "Maps to" dropdown. Solution: As mentioned above, this is the most common problem. Go to Settings > Manage connections and gateways. Select your new gateway and manually add a new "Data Source" to it. Ensure the server address and database name are spelled exactly as they are in your PBIX file. If there's a typo in one but not the other, Power BI won’t see them as a match.
• Issue: The data refresh fails after I switched the gateway. Solution: This usually points to a connectivity or credentials problem.
• Issue: Power BI says the "Gateway is offline or cannot be reached." Solution: Log in to the machine where your new gateway is installed. Open the "On-premises data gateway" application and check its status. The service might have stopped or crashed. Restarting the service often solves the issue. Also, ensure the server itself hasn’t lost network connectivity.

Best Practices for Managing Your Power BI Gateways

To avoid fire drills and make your BI operations more resilient, follow a few best practices for a healthy gateway setup:

• Use Gateway Clusters for High Availability: For any production-level reporting, always install your on-premises gateway on at least two servers and create a "cluster." Power BI will automatically load balance requests between them. If one server goes down for maintenance or fails, your data refreshes will continue uninterrupted through the other active gateway member.
• Separate Dev and Prod Gateways: Never use the same gateway cluster for both development and production workloads. A heavy, unoptimized query from a report in development could slow down or crash the gateway, affecting your business-critical production reports. Allocate separate, dedicated gateways for each environment.
• Keep Your Gateway Software Updated: Microsoft regularly releases updates that include crucial security patches, performance improvements, and bug fixes. Plan brief maintenance windows to keep your gateway software on the latest version.
• Document Your Setup: Keep a simple record of which server hosts which gateway, what datasets rely on it, and who the administrators are. When an issue arises, you'll know exactly where to look instead of trying to reverse-engineer your setup during a crisis.

Final Thoughts

Making a change to a Power BI gateway connection is a simple re-mapping process in the dataset settings, but it depends entirely on thorough preparation. As long as you ensure your new gateway is online and correctly configured with the new data source beforehand, the switch itself will be smooth and seamless.

Managing data infrastructure, from gateways to ETL pipelines, is a critical part of data analysis that often feels like it gets in the way of finding actual insights. That’s a big reason we created Graphed. We wanted to make connecting data sources as easy as a few clicks and a simple question. We automate tying your marketing and sales platforms (like Google Analytics, Shopify, Facebook Ads, and Salesforce) together so you can use natural language to instantly build the dashboards you need - without wrestling with IT configurations or spending your day in technical settings.