showuon commented on code in PR #12561:
URL: https://github.com/apache/kafka/pull/12561#discussion_r978567165
##########
connect/runtime/src/test/java/org/apache/kafka/connect/runtime/distributed/WorkerCoordinatorIncrementalTest.java:
##########
@@ -517,13 +517,13 @@ public void testTaskAssignmentWhenWorkerBounces() {
leaderAssignment = deserializeAssignment(result, leaderId);
assertAssignment(leaderId, offset,
Collections.emptyList(), 0,
- Collections.emptyList(), 0,
+ Collections.emptyList(), 1,
Review Comment:
Wow, nice catch!
> At this point, I think we may want to split this into two separate PRs
that get merged together. We can revert the canRevoke flag from this one, and
then add a downstream PR that fixes how we calculate task-balancing revocations
in tricky situations like when lost or newly-created tasks have just been
assigned. That should fully address cases like the one tested
[here](https://github.com/apache/kafka/blob/cda5da9b65f78b51cdfe5371f712a0d392dbdb4d/connect/runtime/src/test/java/org/apache/kafka/connect/runtime/distributed/WorkerCoordinatorIncrementalTest.java#L427).
Agree! Let's revert the canRevoke flag and then deal with tricky cases in a
follow-up PR. Thank you.
--
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]
