What does a minimum-viable DLQ configuration need, and why does each setting matter?

At minimum: error tolerance set to not fail the whole pipeline on a single bad record (errors.tolerance=all), a clearly named DLQ topic per source (errors.deadletterqueue.topic.name), context headers enabled so each DLQ record carries the original topic/partition and exception details (errors.deadletterqueue.context.headers.enable=true), and error logging enabled with message content included for emergency visibility (errors.log.enable=true, errors.log.include.messages=true). Skip the headers and debugging a DLQ later becomes guesswork.

Expected key concepts: errors.tolerance, errors.deadletterqueue.topic.name, context.headers.enable, errors.log.enable, DLQ configuration, Kafka Connect, debugging headers