Forum Discussion
Databricks connect with Azure SQL Database
If the successful cell only assigns the connection string and properties, it has not tested a database connection. The failure during spark.read.jdbc is therefore consistent with configuration succeeding while the first real connection fails.
Start by expanding the exception to its underlying SQLServerException, including the SQL Server error number and message. HV000 alone does not establish an authentication mismatch.
Next, verify the JDBC driver version on the Databricks compute and use Microsoft's SQL Server JDBC driver for Azure SQL. Compare databaseName, server, and authentication properties with the working client. A personal account label does not specify whether SQL password or Microsoft Entra authentication is being used.
Test a minimal SELECT 1 query from that same compute, then check its network path and database permissions. Share the sanitized connection options, runtime version, and nested exception; those distinguish connection failure from table-access failure without guessing.