Connections

ClueCDC supports PostgreSQL and MySQL as sources and destinations, and AWS S3 and MinIO as object-storage destinations. Kafka and Kafka Connect clusters are registered infrastructure resources. Object storage uses the Aiven S3 sink and JSONL files.

Connection is the only persisted endpoint model. Sources and destinations are capability-filtered views of that resource; pipelines and deliveries refer to the same connection UUID. Credentials are write-only: public APIs return masked credential status. Kafka Connect receives references resolved through a private machine-authenticated API; plaintext credentials remain encrypted in metadata.

Test every connection from the ClueCDC runtime network. localhost inside a container means that container, not the host.