SQL Insert

Inserts a row into an SQL database for each message, and leaves the message content unchanged.

Field	Type	Default	Description
Driver	string	—	Database driver (required)
DSN	string	—	Connection string for the target database (required)
Table	string	—	The target table (required)
Columns	array	—	Column names to populate (required)
Args Mapping	Bloblang	—	Maps message fields to column values. Must evaluate to an array matching the number of columns (required)
Prefix	string	—	An optional prefix to prepend before INSERT
Suffix	string	—	An optional suffix to append (e.g., ON CONFLICT (name) DO NOTHING)
Init Statement	string	—	SQL statement to execute on the first connection
Conn Max Idle Time	string	—	Maximum idle duration before a connection is closed
Conn Max Life Time	string	—	Maximum total lifetime of a connection
Conn Max Idle	integer	2	Maximum number of idle connections in the pool
Conn Max Open	integer	0	Maximum number of open connections (0 = unlimited)

Supported drivers: mysql, postgres, clickhouse, mssql, sqlite, oracle, snowflake, trino, gocosmos, spanner.

Conflict Handling

Use the Suffix field to handle conflicts. For example, in PostgreSQL you can use ON CONFLICT (id) DO NOTHING to skip duplicate rows, or ON CONFLICT (id) DO UPDATE SET name = EXCLUDED.name to upsert.

tip

When using the SQL Insert processor in a pipeline (rather than as an output), the message content remains unchanged after the insert. This allows you to chain additional processors or route the message to other outputs after the database insert.