SQL Migration

SQL Migration imports tables from a relational database into MongoDB. It's a Pro feature, included in the free Pro key.

Supported sources

PostgreSQL, MySQL, MariaDB, Microsoft SQL Server and Oracle. The drivers are bundled; Oracle runs in thin mode, so no Oracle client is needed.

For other databases (IBM Db2, SAP ASE, SQLite), export the tables to CSV and use Import.

Migrate step by step

1. Add a source. Save the SQL connection: host, port, database, user and password (encrypted), and a TLS mode of disable, require or verify with an optional CA file. The source is always read in a read-only session.
2. Introspect the schema. MotionQL lists tables, columns, types, primary keys and foreign keys.
3. Map each table to a collection. Choose columns, filters and type conversions.
4. Model relationships. For many-to-one and one-to-many relationships, choose:
   • Reference: keep the foreign key as a field, like customerId. Best when the related data is large or changes on its own.
   • Embed: nest the related rows inside the document, like an items array inside each order. Best when you always read them together.
5. Preview the documents the mapping will produce.
6. Run. Progress shows rows read, written and failed per table, and you can cancel.

AI relationship suggestions

On a connection with Allow AI on, the assistant can suggest which relationships to embed or reference. It only sees table and column names, types and keys, never row data.

Repeat it

Save a migration as a scheduled task, for example to refresh a MongoDB copy of a reporting database every night.

The other direction

To move MongoDB data into SQL, use Export → SQL to write INSERT statements for generic SQL, MySQL or PostgreSQL. A task can write one script per collection on a schedule.