apache/kafka · warning · IllegalArgumentException

Partition {partition} was not included in the original reque

Error message

Partition {partition} was not included in the original request

What it means

Thrown by DeleteConsumerGroupOffsetsResult.partitionResult when the caller asks for the per-partition result of a partition that was not in the original Admin.deleteConsumerGroupOffsets request set. The result object only tracks partitions the user originally passed; asking for any other TopicPartition is a programming error and is rejected immediately (IllegalArgumentException) rather than via the future.

Source

Thrown at clients/src/main/java/org/apache/kafka/clients/admin/DeleteConsumerGroupOffsetsResult.java:47

 * The result of the {@link Admin#deleteConsumerGroupOffsets(String, Set)} call.
 */
@InterfaceAudience.Public
public class DeleteConsumerGroupOffsetsResult {
    private final KafkaFuture<Map<TopicPartition, Errors>> future;
    private final Set<TopicPartition> partitions;


    DeleteConsumerGroupOffsetsResult(KafkaFuture<Map<TopicPartition, Errors>> future, Set<TopicPartition> partitions) {
        this.future = future;
        this.partitions = partitions;
    }

    /**
     * Return a future which can be used to check the result for a given partition.
     */
    public KafkaFuture<Void> partitionResult(final TopicPartition partition) {
        if (!partitions.contains(partition)) {
            throw new IllegalArgumentException("Partition " + partition + " was not included in the original request");
        }
        final KafkaFutureImpl<Void> result = new KafkaFutureImpl<>();

        this.future.whenComplete((topicPartitions, throwable) -> {
            if (throwable != null) {
                result.completeExceptionally(throwable);
            } else if (!maybeCompleteExceptionally(topicPartitions, partition, result)) {
                result.complete(null);
            }
        });
        return result;
    }

    /**
     * Return a future which succeeds only if all the deletions succeed.
     * If not, the first partition error shall be returned.
     */
    public KafkaFuture<Void> all() {

View on GitHub (pinned to c31c9215e1)

Solutions

1. Pass the exact same TopicPartition object (or an equal topic+partition) that you included in the original deleteConsumerGroupOffsets call.
2. Iterate over the original partition set instead of constructing new TopicPartition instances for lookup.
3. Re-issue Admin.deleteConsumerGroupOffsets with the full set of partitions you intend to query.
4. Verify topic name spelling, casing, and partition index before lookup.

Example fix

// before
Set<TopicPartition> req = Set.of(new TopicPartition("orders", 0));
DeleteConsumerGroupOffsetsResult r =
    admin.deleteConsumerGroupOffsets("grp", req);
r.partitionResult(new TopicPartition("orders", 1)).get(); // not in set

// after - lookup only partitions present in the request set
for (TopicPartition tp : req) {
    r.partitionResult(tp).get();
}

// Keep the exact Set<TopicPartition> handed to deleteConsumerGroupOffsets and
// only query partitionResult(...) for members of that set.
Set<TopicPartition> requested = Set.of(
    new TopicPartition("orders", 0),
    new TopicPartition("orders", 1));
DeleteConsumerGroupOffsetsResult result =
    admin.deleteConsumerGroupOffsets(groupId, requested);

TopicPartition query = new TopicPartition("orders", 0);
if (requested.contains(query)) {
    result.partitionResult(query).get();
} else {
    log.warn("{} was not part of the delete request; skipping", query);
}

// Note: the exception is thrown SYNCHRONOUSLY by partitionResult(...), not via the future.
try {
    result.partitionResult(tp).get();
} catch (IllegalArgumentException e) {
    // `tp` was not in the set passed to deleteConsumerGroupOffsets.
    // Either reuse the original set, or drop this partition from processing.
    log.warn("Skipping {}: {}", tp, e.getMessage());
}

When it happens

Trigger: Calling result.partitionResult(tp) with a TopicPartition that is not equal (topic+partition) to one of the entries in the Set<TopicPartition> passed to Admin.deleteConsumerGroupOffsets. Typical when the caller builds the partition set and the lookup key from different sources, or mutates the set between request and result lookup.

Common situations: Stale references (request was built from one partition list, lookup uses another); copying example code and forgetting to substitute the actual request set; off-by-one in partition numbering; using a TopicPartition object whose topic string case differs from the request.

Related errors

• Topic {topic} was not included in the original request
• Topic partition {partition} was not included in the request
• TransactionalId `{transactionalId}` was not included in the
• TransactionalId `{transactionalId}` was not included in the
• Cannot specify a negative version level.

AI-assisted analysis of apache/kafka@c31c9215e1 (2026-08-03). Data as JSON: /data/errors/1145e28d098fdc83.json. Report an issue: GitHub.