Zum Hauptinhalt springen

Databases

The zbsync_sql module connects zSYNC to relational databases via SQLAlchemy. For Snowflake as a data warehouse source, see the dedicated Snowflake page.

SQL​

Supported databases:

  1. MSSQL

  2. MySQL

  3. DB2

  4. PostgreSQL

Before using a SQL Grabber or Dumper, create a SQL Connection (Connections → SQL) with the database type, host, port, credentials and database name.

There are three SQL worker types:

  • SQL Grabber — runs a SELECT query against the connection and returns the result as a list of records, ready for a Mapper or Dumper.
  • SQL Dumper ("Stores data in sql databases") — runs INSERT/UPDATE statements to write records into a table, based on configured key fields.
  • SQL Executor ("Execute SQL") — runs arbitrary, freeform SQL (not restricted to SELECT), useful for UPDATE, DELETE, DDL statements, or any query whose result doesn't need to flow into the rest of the pipeline as structured data.

SQL executor​

Use the Execute SQL worker when you need to run a raw SQL statement that isn't a simple select-and-map, for example cleanup queries, stored procedure calls, or schema/DDL changes.

  • Connection — the SQL Connection to run against.
  • SQL — the raw SQL statement(s) to execute. Can reference upstream pipeline data the same way other zSYNC code fields do.
  • Unlike the SQL Grabber, the executor does not assume a SELECT shape — it is meant for statements executed for their side effects rather than for the data they return.