bowenli86 commented on code in PR #323:
URL:
https://github.com/apache/flink-connector-kafka/pull/323#discussion_r4354162184
##########
flink-connector-kafka/src/main/java/org/apache/flink/connector/kafka/dynamic/source/enumerator/DynamicKafkaSourceEnumerator.java:
##########
@@ -907,21 +969,87 @@ private void reassignReportedSplits() {
.map(reportedSplit -> reportedSplit.split)
.collect(Collectors.toList());
splitAssignmentStrategy.onRecoveredSplits(activeSplits,
enumContext.currentParallelism());
+ activeSplits.forEach(split ->
currentSplitOwners.remove(toPartitionKey(split)));
+
+ if (!activeSplits.isEmpty()) {
+ // The initial gate blocks every reader until all reports are
reconciled, so the
+ // reporting reader only controls a deferred flush. Enqueue one
batch per cluster
Review Comment:
Updated in 039b622a. The comment now says the closed gate makes
`registeredReaders()` empty, so the reader ID has no effect here, and the later
flush covers all readers.
--
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]
