Java, How to get number of messages in a topic in apache kafka

Question

I am using apache kafka for messaging. I have implemented the producer and consumer in Java. How can we get the number of messages in a topic?

Answer 1

Using the Java client of Kafka 2.11-1.0.0, you can do the following thing :

    KafkaConsumer<String, String> consumer = new KafkaConsumer<>(props);
    consumer.subscribe(Collections.singletonList("test"));
    while(true) {
        ConsumerRecords<String, String> records = consumer.poll(100);
        for (ConsumerRecord<String, String> record : records) {
            System.out.printf("offset = %d, key = %s, value = %s%n", record.offset(), record.key(), record.value());

            // after each message, query the number of messages of the topic
            Set<TopicPartition> partitions = consumer.assignment();
            Map<TopicPartition, Long> offsets = consumer.endOffsets(partitions);
            for(TopicPartition partition : offsets.keySet()) {
                System.out.printf("partition %s is at %d\n", partition.topic(), offsets.get(partition));
            }
        }
    }

Output is something like this :

offset = 10, key = null, value = un
partition test is at 13
offset = 11, key = null, value = deux
partition test is at 13
offset = 12, key = null, value = trois
partition test is at 13

Answer 2

It is not java, but may be useful

./bin/kafka-run-class.sh kafka.tools.GetOffsetShell 
  --broker-list <broker>:  <port> 
  --topic <topic-name> --time -1 --offsets 1 
  | awk -F  ":" '{sum += $3} END {print sum}'

Answer 3

You may use kafkatool. Please check this link -> http://www.kafkatool.com/download.html

Kafka Tool is a GUI application for managing and using Apache Kafka clusters. It provides an intuitive UI that allows one to quickly view objects within a Kafka cluster as well as the messages stored in the topics of the cluster.

Answer 4

To get all the messages stored for the topic you can seek the consumer to the beginning and end of the stream for each partition and sum the results

List<TopicPartition> partitions = consumer.partitionsFor(topic).stream()
        .map(p -> new TopicPartition(topic, p.partition()))
        .collect(Collectors.toList());
    consumer.assign(partitions); 
    consumer.seekToEnd(Collections.emptySet());
Map<TopicPartition, Long> endPartitions = partitions.stream()
        .collect(Collectors.toMap(Function.identity(), consumer::position));
    consumer.seekToBeginning(Collections.emptySet());
System.out.println(partitions.stream().mapToLong(p -> endPartitions.get(p) - consumer.position(p)).sum());

Answer 5

If you have access to server's JMX interface, the start & end offsets are present at:

kafka.log:type=Log,name=LogStartOffset,topic=TOPICNAME,partition=PARTITIONNUMBER
kafka.log:type=Log,name=LogEndOffset,topic=TOPICNAME,partition=PARTITIONNUMBER

(you need to replace TOPICNAME & PARTITIONNUMBER). Bear in mind you need to check for each of the replicas of given partition, or you need to find out which one of the brokers is the leader for a given partition (and this can change over time).

Alternatively, you can use Kafka Consumer methods beginningOffsets and endOffsets.