adixitconfluent commented on code in PR #17539:
URL: https://github.com/apache/kafka/pull/17539#discussion_r1826561286
##########
core/src/main/java/kafka/server/share/DelayedShareFetch.java:
##########
@@ -146,18 +149,47 @@ public void onComplete() {
*/
@Override
public boolean tryComplete() {
- topicPartitionDataFromTryComplete = acquirablePartitions();
+ if (anySharePartitionNoLongerExists()) {
+ return forceComplete();
+ }
+ Map<TopicIdPartition, FetchRequest.PartitionData> topicPartitionData =
acquirablePartitions();
Review Comment:
Hi @junrao , we can do that but then once we are in the `else` of this check
-` if (anyTopicIdPartitionHasLogReadError(replicaManagerReadResponse) ||
isMinBytesSatisfied(topicPartitionData))`, after we have release the partitions
lock, we also need to` partitionsToComplete.clear()`. Hence, I've tried to
avoid doing this by assigning `partitionsToComplete` once we are sure that we
can do a `forceComplete()`
--
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]
