how to delete invalid data source credentials for a dataflow

[Anonymous]

 

I have a dataflow has 3 credentials while 2 of them are invalid and stopping dataflow from refreshing. I just want to know how can I delete the invalid credentials.

I've checked the queries in this dataflow, data source comes from only one single sharepoint folder.

the reason of those invalid credentials could be due to previous sharepoint folder links.

[Anonymous · Accepted solution]

Hi, I think this might work.

After changing all your current dataflow to point to valid credentials, do the following:

• From your workspace with the dataflow > select the workflow > ... > 'Export .json'
• Click New > Dataflow > select 'Import Model' and select your exported .json file from previous step.
• A new version of your dataflow is created with a (1) at the end of the name
• Delete the old Dataflow and rename your new dataflow with the same name

The new Dataflow will only have the valid credentials (no errors if you made the changes correctly).  You may have to relink any reports or datasets that use the flows to the new one.

I have not figured out any method for simply removing unwanted connections.

[bfe212]

Hello, I've solved the issue by editing the dataflows and refreshing the faulty sources. I was then prompted to enter a valid credtial and it succeeded. 

[Anonymous]

actually I have done it in a similar way, to re-build the dataflow.

I created a dataflow with blank query and copy&paste queries from old dataflow. exporting json file is an easier way.

[Greg_Deckler]

@Anonymous Use the Edit credentials link to edit the credentials.

[Anonymous]

@Greg_Deckler thank you for reply. I have looked in edit credentials. there is only options and "sign in" button. I don't have the access to this sharepointsiteUrl now, therefore not able to sign in.

since my dataflow doesn't use any datasource from this sharepointsiteUrl anymore, so I just want to delete this credential if ever possible.