The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredCompete to become Power BI Data Viz World Champion! First round ends August 18th. Get started.
Hi Experts,
I'm facing an issue when refreshing my Power BI Desktop report that's connected to a SQL Server view hosted on AWS US WEST.
Sometimes the refresh is very slow, and often it fails with the following error:
OLE DB or ODBC error: [DataSource.Error] Microsoft SQL: A transport-level error has occurred when receiving results from the server. (provider: TCP Provider, error: 0 - An existing connection was forcibly closed by the remote host.)
This also causes dependent tables in the model to fail with:
"Load was cancelled by an error in loading a previous table."
SQL View: ~Millions+ rows returned
Power BI Desktop: July 2025 version
My Location: India
Data Gateway and Power BI Desktop are on the same Windows laptop (77).
Power BI License: Premium per user.
However, my onsite teammate in the US is using the same .pbix file on his laptop:
Internet speed: 1 Gbps
Premium license
Connected to same SQL view via customer gateway , probably enterprise level premium license for the workspace.
Refresh is very fast for him
Every 2 hours once we are doing incremental refresh at powerbi service.
How can I prevent the SQL connection from being forcibly closed during a refresh in Power BI Desktop?
Are there any settings or techniques to improve refresh performance in Power BI Desktop over VPN?
Could license type (Pro vs Premium) or network speed impact refresh behavior this much?
Any guidance, suggestions, or troubleshooting tips would be greatly appreciated.
Thanks in advance!
Manoj Prabhakar
Hi @manoj_0911,
Just following up to see if the solution provided was helpful in resolving your issue. Please feel free to let us know if you need any further assistance.
Best regards,
Prasanna Kumar
Hi @manoj_0911,
Just following up to see if the solutions provided by community members were helpful in addressing the issue.
If one of the responses helped resolve your query, please consider marking it as the Accepted Solution. Feel free to reach out if you need any further clarification or assistance.
Best regards,
Prasanna Kumar
Hi @manoj_0911,
Just following up to see if the solutions provided by community members were helpful in addressing the issue.
If one of the responses helped resolve your query, please consider marking it as the Accepted Solution. Feel free to reach out if you need any further clarification or assistance.
Best regards,
Prasanna Kumar
Hi @manoj_0911,
Thank you for reaching out to the Microsoft Fabric Forum Community. Thanks to @RemyO fpr prompt and helpful response.
Yes, that would be the recommended approach. It helps by:
Using a refreshed dataset hosted in US.
Avoiding VPN-based slow SQL queries.
Building reports faster in India using cached data from the Service.
Thanks & Regards,
Prasanna kumar
how to do this ? "
So publish the dataset. Set your refresh settings
Then connect your Desktop to the refreshed dataset.
"
1) You cant prevent it. Your connection may be to slow or unstable, hitting time out limits.
2) import less data Do not import unused columns and/or rows
3) maybe and yes. Also the speed of your SQL server might impact it. And whether you import from a view or tabel (table is better)
But why dont you set a refresh from service running from US ? So publish the dataset. Set your refresh settings
Then connect your Desktop to the refreshed dataset.
This would be the best solution to your performance because you dont have to refresh anymore
User | Count |
---|---|
82 | |
81 | |
36 | |
32 | |
32 |
User | Count |
---|---|
93 | |
79 | |
62 | |
54 | |
51 |