dajac commented on code in PR #14271:
URL: https://github.com/apache/kafka/pull/14271#discussion_r1307484913
##########
group-coordinator/src/main/java/org/apache/kafka/coordinator/group/GroupCoordinatorService.java:
##########
@@ -426,9 +429,43 @@ public CompletableFuture<ListGroupsResponseData>
listGroups(
return
FutureUtils.failedFuture(Errors.COORDINATOR_NOT_AVAILABLE.exception());
}
- return FutureUtils.failedFuture(Errors.UNSUPPORTED_VERSION.exception(
- "This API is not implemented yet."
- ));
+ List<CompletableFuture<ListGroupsResponseData>> futures = new
java.util.ArrayList<>(Collections.emptyList());
+ for (int i = 0; i < numPartitions; i++) {
+ futures.add(runtime.scheduleReadOperation("list_groups",
+ new TopicPartition(Topic.GROUP_METADATA_TOPIC_NAME, i),
+ (coordinator, __) -> coordinator.listGroups(context,
request)
Review Comment:
Basically
[here](https://github.com/apache/kafka/pull/14271/files#diff-00f0f81cf13e66781777d94f7d2e68a581663385c37e98792507f2294c91bb09R418),
you need to use `lastCommittedOffset` to ensure that you only read committed
state. I don't know if we support the Stream API but the other APIs should
accept an argument called `epoch`. You can use `lastCommittedOffset` as the
`epoch.`
--
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]
