squah-confluent commented on code in PR #21692:
URL: https://github.com/apache/kafka/pull/21692#discussion_r2915953508
##########
group-coordinator/src/main/java/org/apache/kafka/coordinator/group/OffsetMetadataManager.java:
##########
@@ -630,23 +631,31 @@ public CoordinatorResult<OffsetCommitResponseData,
CoordinatorRecord> commitOffs
.setPartitionIndex(partition.partitionIndex())
.setErrorCode(Errors.OFFSET_METADATA_TOO_LARGE.code()));
} else {
+ // Resolve topic ID if it's ZERO_UUID
+ Uuid resolvedTopicId = topic.topicId();
+ if (resolvedTopicId.equals(Uuid.ZERO_UUID)) {
+ resolvedTopicId = groupMetadataManager.image()
+ .topicMetadata(topic.name())
+ .map(CoordinatorMetadataImage.TopicMetadata::id)
+ .orElse(Uuid.ZERO_UUID);
+ }
// Validate commit per-partition
validator.validate(
topic.name(),
- topic.topicId(),
+ resolvedTopicId,
partition.partitionIndex()
);
log.debug("[GroupId {}] Committing offsets {} for
partition {}-{}-{} from member {} with leader epoch {}.",
- request.groupId(), partition.committedOffset(),
topic.topicId(), topic.name(), partition.partitionIndex(),
+ request.groupId(), partition.committedOffset(),
resolvedTopicId, topic.name(), partition.partitionIndex(),
request.memberId(), partition.committedLeaderEpoch());
topicResponse.partitions().add(new
OffsetCommitResponsePartition()
.setPartitionIndex(partition.partitionIndex())
.setErrorCode(Errors.NONE.code()));
final OffsetAndMetadata offsetAndMetadata =
OffsetAndMetadata.fromRequest(
- topic.topicId(),
+ resolvedTopicId,
Review Comment:
We shouldn't bake the topic id when the offset commit specifies a topic name
only. This impacts behavior when topics are recreated.
```suggestion
topic.topicId(),
```
##########
group-coordinator/src/main/java/org/apache/kafka/coordinator/group/OffsetMetadataManager.java:
##########
@@ -630,23 +631,31 @@ public CoordinatorResult<OffsetCommitResponseData,
CoordinatorRecord> commitOffs
.setPartitionIndex(partition.partitionIndex())
.setErrorCode(Errors.OFFSET_METADATA_TOO_LARGE.code()));
} else {
+ // Resolve topic ID if it's ZERO_UUID
+ Uuid resolvedTopicId = topic.topicId();
+ if (resolvedTopicId.equals(Uuid.ZERO_UUID)) {
+ resolvedTopicId = groupMetadataManager.image()
+ .topicMetadata(topic.name())
+ .map(CoordinatorMetadataImage.TopicMetadata::id)
+ .orElse(Uuid.ZERO_UUID);
+ }
// Validate commit per-partition
validator.validate(
topic.name(),
- topic.topicId(),
+ resolvedTopicId,
partition.partitionIndex()
);
log.debug("[GroupId {}] Committing offsets {} for
partition {}-{}-{} from member {} with leader epoch {}.",
- request.groupId(), partition.committedOffset(),
topic.topicId(), topic.name(), partition.partitionIndex(),
+ request.groupId(), partition.committedOffset(),
resolvedTopicId, topic.name(), partition.partitionIndex(),
Review Comment:
We shouldn't bake the topic id when the offset commit specifies a topic name
only. This impacts behavior when topics are recreated.
```suggestion
request.groupId(), partition.committedOffset(),
topic.topicId(), topic.name(), partition.partitionIndex(),
```
--
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]
