Power BI Report Publishing Problem: Duplicate Reports and Semantic Models

[Gandhia]

Hi Power BI Community,

I need help with a Power BI report I'm working on. Here are the details of my issue:

I am creating a Power BI report. I downloaded an older version of the report from the service (by downloading the PBIX) and modified the report as needed. The deployment pipeline includes dev and prod environments. When I publish the report to dev, it does not replace the previous report nor does it use the same semantic model in the lineage view. Instead, it creates entirely new entries (a new semantic model with the same name as the previous one and a report also with the same name). Naturally, when I deploy to prod, the prod environment ends up with two reports in the original data source.

In the previous lineage view, the path was: Smartsheet -> Semantic model -> Report

In this newer version, the path is: Semantic model -> Report

In production, the lineage view for both versions is: Smartsheet -> Semantic model -> Report

So my questions/issues are as follows:

1. What is causing this? Why is this happening?
2. How can I make the report use the exact same data source in the lineage view?

I would greatly appreciate it if someone could help me resolve this issue.

Thank you in advance!

[Anonymous]

Hi @Gandhia ,

When you published the report from desktop to the Dev workspace and tried to deploy it using the pipeline to QA, an extra copy of the pbix and dataset was deployed instead of overwriting the existing version in QA. This issue happens every time you try to use the pipeline.Please publish your pbix files in both dev and prod environments.

Hope it helps!

Best regards,
Community Support Team_ Scott Chang

If this post helps then please consider Accept it as the solution to help the other members find it more quickly.

[Gandhia]

Thank you for your answer. What do you mean "This issue happens every time you try to use the pipeline"? If I am right, it should NOT happen every time? Shouldn't it overwrite previous report and dataset (aka semantic model) with same name? I think it should. 

So you mean in Power BI Desktop I should publish report not just dev but development AND production? Is this legit solution? I think then the idea of the deployment pipeline disappears. 

[Anonymous]

Hi Gandhia
,

It seems you’re facing challenges with maintaining lineage and ensuring consistent semantic models across your Power BI environments. Let’s dive into the issue and explore solutions:

1. Understanding the Behavior: When you publish a report from Power BI Desktop to different environments (dev and prod), Power BI identifies datasets based on their connections. If there are even minor discrepancies, it treats them as separate entities. This behavior explains why you’re seeing new semantic models and reports instead of replacing existing ones.
2. Potential Causes:
• Connection Differences: Even if the data source structure is identical, variations in server names, database names, or authentication methods can lead to separate models.
• Publishing Process: Ensure you’re publishing from the correct PBIX file and that dataset connections match between dev and prod.
3. Resolution Steps: To address this issue and maintain consistent lineage:
• Standardize Connections: Verify that connection details (server names, credentials) are identical across environments. Adjust them manually in the Power BI Service if needed.
• Check Gateway Settings: If using an on-premises data gateway, ensure consistent configuration across environments.
• Re-publish Carefully: Republish the report, ensuring the PBIX file used for both dev and prod reflects the correct connections.
• Remove Unused Models: In the Power BI Service, delete any redundant semantic models to avoid confusion. Navigate to the Datasets section and remove duplicates.
• Consult Documentation: Microsoft’s documentation provides best practices for deploying Power BI reports. Review it for additional insights.
4. Best Practices:
• Version Control: Use Git or other source control to manage changes and maintain consistency.
• Thorough Testing: Before deployment, rigorously test in dev to catch any discrepancies.

By following these steps and adhering to best practices, you’ll resolve the issue of duplicated semantic models and reports in your Power BI environments. 🚀

Feel free to reach out if you need further assistance! 😊

Regards,

Chiranjeevi Kudupudi

[Gandhia]

Thank you for your comprehensive response, I greatly appreciate it!! However, I didn't get help from that, at least not something concrete that would help me fix the problem. What differences can there be in the data source? How can I find out the details of the previous data source, such as authentication methods?

By the way, I noticed that Power BI has for some reason created a new semantic model, as it has assigned me as the owner of the semantic model (my colleague still owns the original semantic model). Does this help at all? What other additional information could I provide?

How can I proceed or how can I fix this issue?