Problem Statement

JDBC Sink Connectors for PostgreSQL complains below when parsing an array type:

Caused by: org.apache.kafka.connect.errors.ConnectException: Unsupported type for column value: ARRAY
at com.redpanda.kafka.connect.jdbc.dialect.GenericDatabaseDialect.formatColumnValue(GenericDatabaseDialect.java:1718)
at com.redpanda.kafka.connect.jdbc.dialect.GenericDatabaseDialect.writeColumnSpec(GenericDatabaseDialect.java:1643)

 

Solution

JDBC Sink Connectors for PostgreSQL supports the following types in an array:

INT8
INT16
INT32
INT64
FLOAT32
FLOAT64
BOOLEAN
STRING

If a type in your array is none of the above, the Kafka Connector can't understand it.

Possible workarounds are:

• changing the message structure to use arrays of primitives only, or
• trying to use using Single Message Transforms (for example Flatten) to get rid of the complex array right before the message gets processed by the connector.

 

If you have any further questions, or need further guidance, please visit support.redpanda.com