Don't miss your chance to take exam DP-600 or DP-700 on us!
Request nowLearn from the best! Meet the four finalists headed to the FINALS of the Power BI Dataviz World Championships! Register now
Hello,
the Refresh in Power BI Service is not working. I always get an error: Unable to refresh the model (id=10188460) because it references an unsupported data source.
The table where I transform the data with different python scripting cause this error. When I disable loading this table it works again.
I set all data source to public according to the microsoft : https://learn.microsoft.com/en-us/power-bi/connect-data/desktop-python-in-query-editor
But it still doesnt work.
Any tips?
Greetings
Yannick
Solved! Go to Solution.
Hi. Please help us understand the datasource. Are you running python as the first step for getting data? I mean using the get data from python script. Or are you getting data from a source and then runing a py script as a transformation? if it's this second option please share with us which the datasource is, beacase the source could be related to the issue instead of python.
You can also share the python transformation, maybe it's not necessary to use it and we can find an alternative of the transformation with Power Query.
Please check out what Ibendlin is saying. Running Python at that level must configure an On Premise Data Gateway in personal mode to run a refresh at Power Bi Service.
I hope that helps,
Happy to help!
Hello,
thanks for your replies. The Data Source is a SQL Databse. Then I run 3 different python scripts and a lot of different Query Steps and Merges to get my result.
I have different SQL Procedures. They are like this: Insert column a to table a from column a from table b. (more difficult of course) and I need to know where column a is coming from. This was quite tough. It tooks me days to achieve this. I Dont think it s doable just with M Code.
So there is no other way then use a personal gateway?
Yes. The gateway is required. The doc you where reading specify the following:
I can't agree more with Ibendlin. You should try solving the transformations sooner. The sooner the better.
If you just need to do it on python, there might be an alternative for the gateway. However, it might involve additional costs depending on how much you use it. It would be something like this: https://community.fabric.microsoft.com/t5/Community-Blog/Web-Source-to-Avoid-Power-BI-Gateway-with-A...
I hope that helps,
Happy to help!
You should first attempt to do this in SQL. Then in M (everything can be done in M) . Running Python scripts in Power Query should be your very last option. And yes, it absolutely requires a personal gateway if you want to refresh in the service.
Yeah you re probably right. Thanks for your reply 🙂
Hi. Please help us understand the datasource. Are you running python as the first step for getting data? I mean using the get data from python script. Or are you getting data from a source and then runing a py script as a transformation? if it's this second option please share with us which the datasource is, beacase the source could be related to the issue instead of python.
You can also share the python transformation, maybe it's not necessary to use it and we can find an alternative of the transformation with Power Query.
Please check out what Ibendlin is saying. Running Python at that level must configure an On Premise Data Gateway in personal mode to run a refresh at Power Bi Service.
I hope that helps,
Happy to help!
Python scripts require the use of a Personal Gateway.
Share feedback directly with Fabric product managers, participate in targeted research studies and influence the Fabric roadmap.
Check out the February 2026 Power BI update to learn about new features.
| User | Count |
|---|---|
| 20 | |
| 18 | |
| 11 | |
| 9 | |
| 7 |
| User | Count |
|---|---|
| 42 | |
| 38 | |
| 21 | |
| 19 | |
| 17 |