frankvicky commented on code in PR #18737:
URL: https://github.com/apache/kafka/pull/18737#discussion_r1955743945
##########
clients/src/main/java/org/apache/kafka/clients/consumer/internals/events/ApplicationEventProcessor.java:
##########
@@ -206,13 +206,23 @@ public void process(ApplicationEvent event) {
}
private void process(final PollEvent event) {
+ // Trigger a reconciliation that can safely commit offsets if needed
to revoke partitions,
+ // as we're processing before any new fetching starts in the app thread
+
requestManagers.consumerMembershipManager.ifPresent(consumerMembershipManager ->
+ consumerMembershipManager.maybeReconcile(true));
if (requestManagers.commitRequestManager.isPresent()) {
- requestManagers.commitRequestManager.ifPresent(m ->
m.updateAutoCommitTimer(event.pollTimeMs()));
+ CommitRequestManager commitRequestManager =
requestManagers.commitRequestManager.get();
+ commitRequestManager.updateTimerAndMaybeCommit(event.pollTimeMs());
+ // all commit request generation points have been passed,
+ // so it's safe to notify the app thread could start the next poll
cycle
+ event.future().complete(null);
requestManagers.consumerHeartbeatRequestManager.ifPresent(hrm -> {
hrm.membershipManager().onConsumerPoll();
hrm.resetPollTimer(event.pollTimeMs());
});
} else {
+ // unblock if commitRequestManager does not exist
+ event.future().complete(null);
Review Comment:
Without `commitRequestManager`, there is no risk of auto-commit request
generation. Additionally, shareConsumer has no auto-commit mechanism, so it's
safe to unblock here.
--
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: jira-unsubscr...@kafka.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
