MetricSign
EN|NLRequest Access
Medium severityconfiguration

Power BI Refresh Error:
DF-Synapse-InvalidDatabaseType

What does this error mean?

The Synapse linked service is connected to an incompatible database type — for example, a serverless pool endpoint where a dedicated pool is required, or a Spark database instead of a SQL pool.

Common causes

  • 1The linked service server endpoint points to the serverless SQL pool (<workspace>-ondemand.sql.azuresynapse.net) but the data flow requires a dedicated pool for write operations
  • 2The database in the dataset refers to a Spark database or Lake database rather than a dedicated SQL pool database
  • 3The linked service was configured for one pool type and then the database was migrated to a different pool tier without updating the linked service

How to fix it

  1. 1Open the Synapse Analytics linked service in ADF Studio and verify the server endpoint.
  2. 2For dedicated pool operations (reads and writes from data flows): the endpoint must be <workspace>.sql.azuresynapse.net — not the -ondemand variant.
  3. 3For serverless pool queries (read-only SELECT queries): use <workspace>-ondemand.sql.azuresynapse.net — but Mapping Data Flow sinks cannot write to serverless pools.
  4. 4If writing to Synapse from a data flow, confirm the target is a dedicated SQL pool — serverless pools are read-only and do not support writes from data flows.
  5. 5In Synapse Studio, verify the database appears under SQL pools > Dedicated SQL pools, not under Apache Spark pools.
  6. 6After correcting the endpoint and database, click 'Test connection' on the linked service and re-run the data flow.

Frequently asked questions

What is the difference between a Synapse dedicated pool and serverless pool?

Dedicated SQL pools are provisioned (fixed DWU) and support reads and writes. Serverless pools are on-demand query engines over storage — they cannot be written to from data flows. Endpoints: dedicated = <workspace>.sql.azuresynapse.net; serverless = <workspace>-ondemand.sql.azuresynapse.net.

Can I use a Synapse Spark pool as a data flow sink?

No — Mapping Data Flow sinks connect to SQL endpoints, not Spark pools. To write to a Spark table, use an Azure Databricks linked service or write to ADLS Gen2 and register the files as Spark tables.

How do I verify which type of database my linked service is connected to?

In Synapse Studio, go to Data hub > Workspace. Dedicated pool databases appear under SQL Pools; serverless under 'Built-in'; Spark under Apache Spark pools. Match your linked service database name against these.

Will downstream Power BI datasets be affected?

Yes — the Synapse write fails and no data reaches the target table. Dependent datasets serve stale figures.

Official documentation: https://learn.microsoft.com/en-us/azure/data-factory/data-flow-troubleshoot-guide

Other configuration errors