I'm new to Azure and its ML Designer. I have an Azure SQL Database set up with my Client IP allowed through the Firewall. I can connect using SSMS without issue using the server path and credentials Azure supplied.
I have set up a Machine Learning Resource and launched the Learning Studio.
Inside the Learning Studio I created a DataStore pointing to my Azure SQL Database and everything seems just fine:
However, when I try to create a DataSet from that DataStore I am getting Login errors:
Login error when creating DataSet
In "More details" I see the following:
ScriptExecutionException was caused by DatabaseConnectionException.\n DatabaseConnectionException was caused by AuthenticationException.\n 'MSSQl' encountered connection erro with error code '18456' while opening connection to server ([REDACTED]), database ([REDACTED]). Either the database name is not valid or username and password is wrong. Exception class: '14'. HResult: x80131904.\n Failed due to inner exception of type: SqlException\n
I am sure the database name and login credentials are correct i.e. they work just fine when connecting from my SSMS. What else can this be? I've checked the Azure SQL Database profile - is there something here regarding security or access that I should be setting?
** UPDATE ** Solved this by going to SQL Server > Active Directory admin. There was nothing set here so I clicked Set Admin and then from the right panel scrolled down to the ML service I had set up. Once I saved it and then created a Dataset in ML I could see the data Preview just fine.