lnbest0707 commented on code in PR #276:
URL:
https://github.com/apache/flink-connector-kafka/pull/276#discussion_r3470492923
##########
flink-connector-kafka/src/main/java/org/apache/flink/connector/kafka/dynamic/source/reader/DynamicKafkaSourceReader.java:
##########
@@ -592,6 +595,12 @@ private void completeAndResetAvailabilityHelper() {
this.availabilityHelperSize = clusterReaderMap.size());
syncAvailabilityHelperWithReaders();
+ if (clusterReaderMap.isEmpty()) {
Review Comment:
Checking only `clusterReaderMap.isEmpty()` may not be sufficient. IIUC, the
sub-readers map is always created as long as the kafka metadata map is not
empty.
The failing case described in the ticket is more related the number of
active assigned splits. When the last active assigned split is removed from the
reader, we should mark it idle. While in such case, this `clusterReaderMap` may
not be empty and code would not fall in this branch.
--
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]
