Hi @bhalicki ,
Thank you for reaching out to Microsoft Community.
When working with Power BI or similar tools, it’s essential to ensure that both the shared semantic model and the local model are compatible with each other. First, verify that both models are using the same version of the semantic model or schema. Any discrepancies between the two models can introduce issues that may affect how the report saves or loads. Even seemingly minor changes in one model such as changes to the schema, measures, or relationships could lead to problems when trying to save or publish the report.
One common issue that could cause the save functionality to break is a conflict with permissions or access settings. It’s crucial to ensure that both models (the shared and local ones) are accessible, and that you have the appropriate permissions to save or publish the report. Additionally, you should check for any restrictions within your environment, such as limitations on cloud storage or on-premises access, that might prevent the save operation from succeeding.
Another potential problem is a conflict between the shared and local models themselves. If both models have overlapping names, keys, or relationships, this can lead to instability when trying to save or publish the report. Carefully review the relationships and structure of both models to identify any potential conflicts. Pay close attention to any conflicting names, measures, or calculated columns between the two models, as these can cause errors.
In Power BI, there are also diagnostic tools that can provide more information about what's going wrong. Using features like the Diagnostic Logs or enabling Trace can help uncover any underlying issues that are preventing the report from being saved successfully. These tools might reveal issues such as conflicting queries, failed data refreshes, or errors in model loading.
If you're trying to save the report to a server or cloud location and encountering issues, it may be worth trying to save the report to your local machine instead. This can help you avoid potential network-related issues or server permission problems that could be interfering with the save process.
Another workaround to consider is using the Save As feature to save the report under a new name. Instead of overwriting the existing report, try saving it as a new file. This can help sidestep issues that might be related to the original report’s settings, cache, or underlying model configuration.
If integrating the local model directly into your existing report continues to cause issues, consider creating a separate report that only uses the local model. This way, you can build the report using just the local model and then link it to the shared semantic model report. You could reference the new report via a dashboard or embed it into the shared report, which would isolate the local model and reduce the risk of conflicts between the two models.
Hope this helps
Best Regards.
Chaithra E.
