Debugging Power BI Web API Refresh Issues

[qmestu]

I have a Power BI report that retrieves a large amount of data from web sources (specifically an API). When I refresh the report, I only see a dialog showing all tables in an “Evaluating” state. Is there a way to see what Power BI is doing in the background during the refresh, or to identify where the process is failing?

I know the API itself is working correctly, since I can successfully test it in Postman and receive valid responses.

[v-lgarikapat]

Hi @qmestu Thank you for reaching out to the Fabric Community.

@AshokKunwar , @danextian , @GeraldGEmerick , Thank you for your prompt responses.

I wanted to follow up and confirm whether you’ve had the opportunity to review the information  shared by @AshokKunwar , @danextian , @GeraldGEmerick  If you have any questions or need further clarification, please don’t hesitate to reach out.

We truly appreciate your engagement and thank you for being an active part of the community.

Best Regards,

Lakshmi.

[AshokKunwar]

Hii @qmestu 

Subject: [SOLVED] Debugging API Refresh: How to Fix "Evaluating" Hangs in Power BI

​The Challenge:

When a Power BI report hangs in the "Evaluating" state during a web API refresh, it’s often because the Mashup Engine is trying to determine the schema or handle authentication before the first row of data is even processed. While Postman proves the API works, it doesn't account for how Power BI’s data privacy settings or connection limits interact with that source.

​The Solution:

You can "peek" under the hood of the refresh process by using Diagnostic Tracing and the Performance Analyzer.

​1. Enable Power Query Diagnostics

​To see exactly what the engine is doing during that "Evaluating" phase:

1. ​In Power BI Desktop, go to File > Options and settings > Options.
2. ​Under the Global section, select Diagnostics.
3. ​Check Enable tracing and click Start diagnostic tracing.
4. ​Run your refresh again. Once it hangs or fails, click Open traces folder to view the detailed .log files showing every HTTP request Power BI attempted.

​2. Isolate the "Lazy Evaluation"

​Power BI uses "Lazy Evaluation," meaning it won't pull data until it's absolutely needed for a visual.

• ​The Fix: Open the Performance Analyzer, click Start Recording, and then Refresh Visuals.
• ​This will show you if the delay is in the DAX Query (calculating the result) or the Other category (waiting for the Web API to return the data).

​3. Common Fixes for API "Evaluating" Hangs

• ​Data Privacy Levels: Go to Options > Current File > Privacy and set it to "Ignore the Privacy Levels". Often, the "Evaluating" state is just Power BI checking if it’s safe to combine data from different API endpoints.
• ​Disable Background Preview: Go to Options > Current File > Data Load and uncheck "Allow data preview to download in the background". This prevents Power BI from trying to refresh all tables simultaneously, which can hit API rate limits.

​Why this works:

​Enabling tracing reveals the raw HTTP status codes (like 429 Too Many Requests or 500 Internal Server Error) that the standard UI hides behind the "Evaluating" label. By adjusting privacy and background load settings, you reduce the "chatter" between Power BI and your API.

​If these debugging steps helped you identify your API bottleneck, please mark this as an "Accepted Solution" to help others!

Best regards,

Vishwanath 

[danextian]

Hi @qmestu 

Without using the diagnostic tool, the latency is likely due to the volume of data. The API may be throttling requests to protect overall performance. Querying large datasets directly through the web connector is generally not recommended.

A better approach is to use dataflows to separate historical and current data, then recombine them in Power Query. Only the current data needs frequent refreshes. If connecting to the historical dataflow is still slow, consider further splitting it into smaller time-based partitions. This reduces repeated API calls and prevents Power Query from retrieving the full dataset on every refresh.

Some organizations that retrieve data from APIs store it in a database and periodically update it with newly created or modified records.

[qmestu]

How does one go about using dataflows?

[GeraldGEmerick]

@qmestu In Power Query Editor, you can turn on diagnostics from the Tools menu.