...

MirrorMaker 2.0 (KIP-382) needs to know the downstream offsets of replicated records in order to provide cross-cluster offset translation. Currently, WorkerSourceTask receives this information from KafkaProducer but throws it away. It's possible that other Connectors may benefit from this change, e.g. see KIP-381, which also proposes to notify SourceTasks of ACK'd records. In particular, this proposal makes it possible to distinguish when records have been durably stored vs when they have been skipped altogether by a SourceConnector.

Public Interfaces

A new The callback method recordLoggedcommitRecord() will be added to SourceTaskoverloaded with an extra parameter:

public abstract class SourceTask implements Task {
---%<---


// existing method
public void commitRecord(SourceRecord sourceRecord) {
  // nop
}


// new method
public void recordLoggedcommitRecord(SourceRecord sourceRecord, RecordMetadata recordMetadata) {
  // nop
}
---%<---

...

• record is ACK'd from producer
• SourceRecord is filtered out from transformation chain, and thus is never sent to Kafka
• SourceTask gives up after retries and skips the SourceRecord.

The new method recordLogged() overloaded version will be invoked only when a record is ACK'd, which implies the record was not filtered and was not skipped. This is somewhat intuitive, as the RecordMetadata can only be filled in by a producer ACK.

After an ACK, WorkerSourceTask currently logs MetadataRecord.offset() and partition() before invoking the commitRecord() callback. I propose to also invoke the new recordLogged() callback overloaded version as well:

task.recordLoggedcommitRecord(preTransformRecord,);

if (recordMetadata != null);
    task.commitRecord(preTransformRecord, recordMetadata);

Compatibility, Deprecation, and Migration Plan

This is a new callback and won't affect existing code beyond an additional no-op function call. 

Rejected Alternatives

• We could introduce a new method name, but overloading an existing method is a lighter touch.
• We could call it recordAckedinvoke commitRecord() , but I think recordLogged() better captures what is being ACK'dwith recordMetadata == null when records are skipped, but this would be surprising and error prone to those using the API.
• SinkRecord has offset and partition fields, so we could potentially use recordLoggedcommitRecord(SinkRecord), but this would be a confusing abuse of SinkRecord's semantics. In particular, it would be weird to have SinkRecord in the SourceTask interface.
• We could add kafkaOffset and kafkaPartition fields to SourceRecord, which would be null prior to ACK and then filled in by WorkerSourceTask after ACK. However, this breaks the "value class" semantics of SourceRecord.
• We could extend SourceRecord with AckedSourceRecord or LoggedSourceRecord, but this seems overkill.
• We could avoid using RecordMetadata in the API, and instead include kafkaOffset, kafkaPartition, kafkaTopic, and kafkaTimestamp as parameters to recordLoggedcommitRecord(). But this is a lot of parameters, and RecordMatadata is already in the clients API.
• We could pass just the kafkaOffset to recordLoggedcommitRecord(), but an offset is mostly meaningless without an associated partition. We could include both kafkaOffset and kafkaPartition, but this doesn't account for transformations that may change the downstream topic name. We could include kafkaTopic as well, but then we might as well include the entire RecordMetadata.