PatrickRen commented on code in PR #19456:
URL: https://github.com/apache/flink/pull/19456#discussion_r851005151
##########
flink-connectors/flink-connector-kafka/src/main/java/org/apache/flink/connector/kafka/source/reader/KafkaPartitionSplitReader.java:
##########
@@ -131,12 +138,7 @@ public RecordsWithSplitIds<ConsumerRecord<byte[], byte[]>>
fetch() throws IOExce
kafkaSourceReaderMetrics.maybeAddRecordsLagMetric(consumer, tp);
}
- // Some splits are discovered as empty when handling split additions.
These splits should be
- // added to finished splits to clean up states in split fetcher and
source reader.
- if (!emptySplits.isEmpty()) {
- recordsBySplits.finishedSplits.addAll(emptySplits);
- emptySplits.clear();
- }
+ markEmptySplitsAsFinished(recordsBySplits);
Review Comment:
The stopping offset of unbounded partitions is set to Long.MAX, so they
won't be treated as empty.
https://github.com/apache/flink/blob/release-1.14.4/flink-connectors/flink-connector-kafka/src/main/java/org/apache/flink/connector/kafka/source/reader/KafkaPartitionSplitReader.java#L311-L331
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
