Monitor semantic model refresh failures and causes in Power BI

[Anonymous]

Hi everyone,

I’m looking for a way to monitor semantic models (datasets) in Power BI, specifically to:

• Identify which semantic models fail during refresh

• Understand the root cause or error messages behind these failures

Ideally, I want a centralized dashboard or tool that can alert me when failures happen and provide insights into what caused them.

I know about the refresh history logs in Power BI Service, but I’m hoping for a more scalable solution, especially if it can support multiple models at once.

Are there any built-in tools, Power BI Premium/Capacity features, APIs, or third-party solutions you’d recommend for this?

Thanks in advance!

[v-tsaipranay]

Hi @Anonymous ,

We haven’t received an update from you in some time. Could you please let us know if the issue has been resolved?
If you still require support, please let us know, we are happy to assist you.

Thank you.

[v-tsaipranay]

Hi @Anonymous ,

I wanted to follow up on our previous suggestions. We would like to hear back from you to ensure we can assist you further.

Also thank you @Poojara_D12 for your insights. Please let us know if there’s anything else we can do to help.

Thank you.

[Poojara_D12]

Hi @Anonymous 

To monitor semantic models (datasets) in Power BI at scale—especially to identify refresh failures and understand their root causes—a more comprehensive and centralized approach is essential than just relying on manual checks in the Power BI Service. If you're using Power BI Premium or Fabric capacities, a highly recommended built-in solution is the Power BI REST API combined with Dataflows or Power Automate to capture refresh history logs across workspaces. Specifically, the Get Refresh History and Get Dataset API endpoints allow you to programmatically retrieve the status, timestamps, and error messages from refreshes. You can automate this via a scheduled Power Automate flow or Azure Logic App, storing the results in a centralized data repository like a SQL database or a Fabric Lakehouse. From there, a Power BI report can be built to create a monitoring dashboard with visual alerts.

Additionally, if you're on Power BI Premium Capacity, the Power BI Premium Capacity Metrics App provides valuable telemetry such as refresh times, failures, memory consumption, and query performance. Although it doesn’t always show detailed failure reasons, it can highlight problematic models over time.

For alerting, Power BI Data Alerts are limited to visuals based on dashboards, so many organizations instead use Power Automate to send email or Teams notifications when a refresh failure is detected via the REST API.

For even more advanced monitoring, third-party tools like PowerBI Sentinel, PowerBI Ops, or Data Mart Pulse offer centralized, enterprise-grade monitoring dashboards. These tools aggregate refresh failures, security changes, and audit logs across all workspaces and offer root cause diagnostics along with proactive alerts.

In short, while refresh history is a starting point, combining the Power BI API with automation tools and optionally third-party solutions enables you to scale monitoring, centralize error tracking, and receive alerts when semantic model refreshes fail.

[v-tsaipranay]

Hi @Anonymous ,

I wanted to check if you had the opportunity to review the information provided. Please feel free to contact us if you have any further questions.

Thank you.

[v-tsaipranay · Accepted solution]

Hi @Anonymous ,

Thank you for reaching out with the Microsoft Fabric Community forum.

Here are some effective methods to centralize monitoring for all your refreshes:

If you are using Fabric or Power BI Premium capacity, the Premium Metrics app provides visibility into refresh queues, run times, and failures for every model within that capacity.

If you need to access information across multiple workspaces, or if you’re using Pro, you can retrieve this data yourself. The Power BI Admin Activity Log (or the Get Refresh History REST API) tracks every refresh attempt and failure. Many customers use a script or Power Automate flow to regularly call this API, store the results in a table, and then create a Power BI report based on that data. By setting a “when failure equals true” condition in Power Automate, you can receive an email or Teams notification as soon as a refresh fails.

Alternatively, few may use ready-made tools like PowerBI Sentinel or an Azure Monitor workbook, which connect to the same logs and offer a dashboard without additional coding.

Hope this helps. Please reach out for further assistance.

Thank you.

[MasonMA]

Great topic! Had similar question before and tried some of the solutions Copilot provided, but none works. Looking forward to seeing a tested approach too.