Editing Dataflow error code "status code: 400"

[Drunken_Sailor]

I have a dataset that will not refresh:

Data source errorUnknown exception. The exception was raised by the IDbCommand interface. Table: all-all-sp-pvp-all

So I then open the dataset on Power Bi Desktop and try to refresh, but get an error that all-all-sp-pvp-all is blocking the refresh of all quiries. So I open power query and see that in all-all-sp-pvp-all  I see this error

ModelNotFoundException: Unknown exception
Details:
Error=Unknown exception

So I go back to the service and try to edit the dataflow that has the table  all-all-sp-pvp-all , but I get this 

I have tried updating my credentials, but that has not helped.  If I refresh it, I do not get any error, it seems fine, but the following dataset cannot seem to read it. How do I troubleshoot this, if I cannot get into it?

[v-kpoloju-msft]

Hi @Drunken_Sailor,

Thanks for reaching out to the Microsoft fabric community forum. After thoroughly reviewing the details you provided, I have identified few workarounds that may help resolve the issue. Please follow these steps:

The issue seems to be a dataset not refreshing properly in Power BI, with a Status code 400 error indicating a possible data connection or authentication problem. a bad request issue, often related to incorrect data structure, permissions, or connection problems.

Please check if the dataflow refreshes successfully in the Power BI Service. Go to Power BI Service > Workspaces > Dataflows, find the dataflow, and manually trigger a refresh. If it fails, check the refresh history for errors. Ensure correct credentials in Dataflow Settings > Data Source Credentials, re-authenticate, and verify access permissions.

Open the Power Query Editor within the Dataflow and check for the presence of the table all-all-sp-pvp-all. Verify the column names, data types, and transformations, and ensure no columns are missing or deleted from the source.

Create a new blank Power BI file and connect directly to the data source, bypassing the dataflow. If the table is absent, the issue is at the source level. If the dataflow is correct, ensure downstream datasets are properly linked. Open the affected dataset in Power BI Service > Settings > Data Source Credentials and confirm it correctly references the dataflow.

The 400 (Bad Request) error may be due to changes in API endpoints (if connecting via API), schema mismatches between the dataflow and dataset, or a potential issue with the Power BI Service. Please check the Power BI Service status page for any updates.

Kindly refer the below following link for better understanding:
Troubleshooting dataflow issues - connection to the data source - Power Query | Microsoft Learn
Understand and optimize dataflows refresh - Power BI | Microsoft Learn

If this post helps, then please give us ‘Kudos’ and consider Accept it as a solution to help the other members find it more quickly.

Thank you.

[Drunken_Sailor · Accepted solution]

Thanks for the response.

I probably was not clear enough, but I had checked that the authentication was correct and also checked trying to open the dataflow to edit it which returned the "something went wrong" message above. So I could not see what was causing the issue becuase I could not open the dataflow on the service itself.

This dataflow (which feeds the dataset for the report) gets aggregated data from a few other dataflows.

I tried exporting it to a json and recreating but got the same issue. So I tried recreating it as brand new, which is when I discovered there was one dataflow that was giving the same error.

I then tried opening the 2nd dataflow I  got the same error. This 2nd dataflow, is one of many that does the same thing (same data structure, same queries, addressing the same APIs) to get batches of data.  I cannot use just one as there is a limit on how much data can be retrieved, but I know I am not reaching the limnit as that just spits out a limit error. The other dataflows worked without issue.  

As I could not edit that second dataflow, I just deleted it, copied one of the working ones, updating the parameters and it now it worked fine.

So it looks like I resolved the issue (still in progress for a refresh, but no immediate errors thrown out) but I lost a lot of time trying to find the issue becuase I could not simply open/edit the dataflows having trouble.

I am glad the dataflow that somehow got corrupted only had a few steps and was easy to rebuild, but I am worried that I have others that are much more complex may end up having the issue and do not have an easy way to fix.  

This enitre report has been running smoothly for months without changes until now.

Also, I did check the Powerr Bi service, but nothing was reported as an issue that would have correlated with the start of the problem.