Lightdash direct-connection boundary¶
Start with the shared setup and dataset. The versioned observations below use the MatrixOne build identified in the overview.
Check the warehouse adapter first¶
Lightdash’s semantic layer connects through supported warehouse adapters. MySQL-protocol compatibility alone is insufficient if the tool has no MySQL adapter.
In the tested 2.573.1 CLI/common/warehouses runtime, supported warehouse types included BigQuery, Databricks, Snowflake, Redshift, PostgreSQL, DuckDB, Trino, ClickHouse and Athena. Both mysql and matrixone were rejected by the native warehouse factory and SQL builder.
Check the supported warehouse list in your installed Lightdash version.
If MatrixOne/MySQL is absent, do not enter MatrixOne’s endpoint as a PostgreSQL warehouse. A PostgreSQL credential object passing a constructor/type check does not demonstrate a PostgreSQL connection to MatrixOne.
Do not publish a
type: mysqlLightdash/dbt configuration as a verified MatrixOne integration unless an actual supported adapter exists and completes queries in your installed version.
No MatrixOne SQL was executed through this Lightdash runtime, and no full Lightdash server/dashboard was deployed in this compatibility check.
Indirect architecture¶
An indirect deployment would extract MatrixOne data into a warehouse supported by both the chosen dbt adapter and Lightdash, define models/metrics there, and connect Lightdash to that warehouse. That architecture requires separate validation of extraction, Decimal precision, incremental updates, refresh timing and metric definitions. It was not tested here and introduces a second query engine; it is not a native MatrixOne connection workaround.
For direct queries to MatrixOne in this tutorial series, start with a guide whose validated path matches your needs, such as Metabase, DataEase or Zeppelin, and review their version-specific limitations.