redpanda-data/connect v4.89.0 on GitHub

For installation instructions check out the getting started guide.

general: PostgreSQL CDC benchmarking suite added with Kafka Connect PostgreSQL benchmark infrastructure and configuration. (@ness-david-dedu, #4216)
parquet_encode: Added configurable timestamp unit support (nanosecond, microsecond, millisecond) to make Parquet output readable by Apache Spark, Databricks, AWS Athena, and DuckDB. (@ankit481, #4294)
oracledb_cdc: Added transaction_id to message metadata. (@josephwoodward, #4328)
oracledb_cdc: Added commit_ts_ms to message metadata. (@josephwoodward, #4331)

confluent: Avro schema reference resolution now handles arbitrary schema shapes and correctly inlines transitive references, fixing misleading errors and missing nested reference resolution. (@twmb, #4247)
mysql_cdc: IAM token refresh and canal recreation before streaming prevents connection failures when snapshots delay binlog streaming. (@josephwoodward, #4295)
oracledb_cdc: Oracle numeric values with missing leading zeros (e.g., '.5') are now normalized to valid JSON format for proper CDC streaming. (@josephwoodward, #4322)

Unreleased

parquet_encode: Added default_timestamp_unit field (values NANOSECOND, MICROSECOND, MILLISECOND) controlling the precision of TIMESTAMP logical types. Default remains NANOSECOND for backwards compatibility. Use MICROSECOND when writing files for Apache Spark/Databricks, AWS Athena or DuckDB, which do not support TIMESTAMP(NANOS). (#3570)

The full change log can be found here.