Using Spring for Apache Kafka

Using Spring for Apache Kafka

Sending Messages

KafkaTemplate

The KafkaTemplate wraps a producer and provides convenience methods to send data to kafka topics. Both asynchronous and synchronous methods are provided, with the async methods returning a Future.

ListenableFuture<SendResult<K, V>> sendDefault(V data);

ListenableFuture<SendResult<K, V>> sendDefault(K key, V data);

ListenableFuture<SendResult<K, V>> sendDefault(int partition, K key, V data);

ListenableFuture<SendResult<K, V>> send(String topic, V data);

ListenableFuture<SendResult<K, V>> send(String topic, K key, V data);

ListenableFuture<SendResult<K, V>> send(String topic, int partition, V data);

ListenableFuture<SendResult<K, V>> send(String topic, int partition, K key, V data);

ListenableFuture<SendResult<K, V>> send(Message<?> message);

Map<MetricName, ? extends Metric> metrics();

List<PartitionInfo> partitionsFor(String topic);

<T> T execute(ProducerCallback<K, V, T> callback);

// Flush the producer.

void flush();

interface ProducerCallback<K, V, T> {

    T doInKafka(Producer<K, V> producer);

}

The first 3 methods require that a default topic has been provided to the template.

The metrics and partitionsFor methods simply delegate to the same methods on the underlying Producer. The executemethod provides direct access to the underlying Producer.

To use the template, configure a producer factory and provide it in the template’s constructor:

@Bean
public ProducerFactory<Integer, String> producerFactory() {
    return new DefaultKafkaProducerFactory<>(producerConfigs());
}

@Bean
public Map<String, Object> producerConfigs() {
    Map<String, Object> props = new HashMap<>();
    props.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, "localhost:9092");
    ...
    return props;
}

@Bean
public KafkaTemplate<Integer, String> kafkaTemplate() {
    return new KafkaTemplate<Integer, String>(producerFactory());
}

The template can also be configured using standard <bean/> definitions.

Then, to use the template, simply invoke one of its methods.

When using the methods with a Message<?> parameter, topic, partition and key information is provided in a message header:

• KafkaHeaders.TOPIC
• KafkaHeaders.PARTITION_ID
• KafkaHeaders.MESSAGE_KEY

with the message payload being the data.

Optionally, you can configure the KafkaTemplate with a ProducerListener to get an async callback with the results of the send (success or failure) instead of waiting for the Future to complete.

public interface ProducerListener<K, V> {

    void onSuccess(String topic, Integer partition, K key, V value, RecordMetadata recordMetadata);

    void one rror(String topic, Integer partition, K key, V value, Exception exception);

    boolean isInterestedInSuccess();

}

By default, the template is configured with a LoggingProducerListener which logs errors and does nothing when the send is successful.

onSuccess is only called if isInterestedInSuccess returns true.

For convenience, the abstract ProducerListenerAdapter is provided in case you only want to implement one of the methods. It returns false for isInterestedInSuccess.

Notice that the send methods return a ListenableFuture<SendResult>. You can register a callback with the listener to receive the result of the send asynchronously.

ListenableFuture<SendResult<Integer, String>> future = template.send("foo");
future.addCallback(new ListenableFutureCallback<SendResult<Integer, String>>() {

    @Override
    public void onSuccess(SendResult<Integer, String> result) {
        ...
    }

    @Override
    public void onFailure(Throwable ex) {
        ...
    }

});

The SendResult has two properties, a ProducerRecord and RecordMetadata; refer to the Kafka API documentation for information about those objects.

If you wish to block the sending thread, to await the result, you can invoke the future’s get() method. You may wish to invokeflush() before waiting or, for convenience, the template has a constructor with an autoFlush parameter which will cause the template to flush() on each send. Note, however that flushing will likely significantly reduce performance.

Receiving Messages

Messages can be received by configuring a MessageListenerContainer and providing a Message Listener, or by using the@KafkaListener annotation.

Message Listeners

When using a Message Listener Container you must provide a listener to receive data. There are currently four supported interfaces for message listeners:

public interface MessageListener<K, V> {} (1)

    void onMessage(ConsumerRecord<K, V> data);

}

public interface AcknowledgingMessageListener<K, V> {} (2)

    void onMessage(ConsumerRecord<K, V> data, Acknowledgment acknowledgment);

}

public interface BatchMessageListener<K, V> {} (3)

    void onMessage(List<ConsumerRecord<K, V>> data);

}

public interface BatchAcknowledgingMessageListener<K, V> {} (4)

    void onMessage(List<ConsumerRecord<K, V>> data, Acknowledgment acknowledgment);

}

1. Use this for processing individual ConsumerRecord s received from the kafka consumer poll() operation when using auto-commit, or one of the container-managed commit methods.
2. Use this for processing individual ConsumerRecord s received from the kafka consumer poll() operation when using one of the manual commit methods.
3. Use this for processing all ConsumerRecord s received from the kafka consumer poll() operation when using auto-commit, or one of the container-managed commit methods. AckMode.RECORD is not supported when using this interface since the listener is given the complete batch.
4. Use this for processing all ConsumerRecord s received from the kafka consumer poll() operation when using one of the manual commit methods.

Message Listener Containers

Two MessageListenerContainer implementations are provided:

• KafkaMessageListenerContainer
• ConcurrentMessageListenerContainer

The KafkaMessageListenerContainer receives all message from all topics/partitions on a single thread. TheConcurrentMessageListenerContainer delegates to 1 or more KafkaMessageListenerContainer s to provide multi-threaded consumption.

KafkaMessageListenerContainer

The following constructors are available.

public KafkaMessageListenerContainer(ConsumerFactory<K, V> consumerFactory,
                    ContainerProperties containerProperties)

public KafkaMessageListenerContainer(ConsumerFactory<K, V> consumerFactory,
                    ContainerProperties containerProperties,
                    TopicPartitionInitialOffset... topicPartitions)

Each takes a ConsumerFactory and information about topics and partitions, as well as other configuration in aContainerProperties object. The second constructor is used by the ConcurrentMessageListenerContainer (see below) to distribute TopicPartitionInitialOffset across the consumer instances. ContainerProperties has the following constructors:

public ContainerProperties(TopicPartitionInitialOffset... topicPartitions)

public ContainerProperties(String... topics)

public ContainerProperties(Pattern topicPattern)

The first takes an array of TopicPartitionInitialOffset arguments to explicitly instruct the container which partitions to use (using the consumer assign() method), and with an optional initial offset: a positive value is an absolute offset by default; a negative value is relative to the current last offset within a partition by default. A constructor forTopicPartitionInitialOffset is provided that takes an additional boolean argument. If this is true, the initial offsets (positive or negative) are relative to the current position for this consumer. The offsets are applied when the container is started. The second takes an array of topics and Kafka allocates the partitions based on the group.id property - distributing partitions across the group. The third uses a regex Pattern to select the topics.

Refer to the JavaDocs for ContainerProperties for more information about the various properties that can be set.

ConcurrentMessageListenerContainer

The single constructor is similar to the first KafkaListenerContainer constructor:

public ConcurrentMessageListenerContainer(ConsumerFactory<K, V> consumerFactory,
                            ContainerProperties containerProperties)

It also has a property concurrency, e.g. container.setConcurrency(3) will create 3 KafkaMessageListenerContainer s.

For the first constructor, kafka will distribute the partitions across the consumers. For the second constructor, theConcurrentMessageListenerContainer distributes the TopicPartition s across the delegateKafkaMessageListenerContainer s.

If, say, 6 TopicPartition s are provided and the concurrency is 3; each container will get 2 partitions. For 5 TopicPartitions, 2 containers will get 2 partitions and the third will get 1. If the concurrency is greater than the number of TopicPartitions, the concurrency will be adjusted down such that each container will get one partition.

Committing Offsets

Several options are provided for committing offsets. If the enable.auto.commit consumer property is true, kafka will auto-commit the offsets according to its configuration. If it is false, the containers support the following AckMode s.

The consumer poll() method will return one or more ConsumerRecords; the MessageListener is called for each record; the following describes the action taken by the container for each AckMode :

• RECORD - commit the offset when the listener returns after processing the record.
• BATCH - commit the offset when all the records returned by the poll() have been processed.
• TIME - commit the offset when all the records returned by the poll() have been processed as long as the ackTime since the last commit has been exceeded.
• COUNT - commit the offset when all the records returned by the poll() have been processed as long as ackCountrecords have been received since the last commit.
• COUNT_TIME - similar to TIME and COUNT but the commit is performed if either condition is true.
• MANUAL - the message listener is responsible to acknowledge() the Acknowledgment; after which, the same semantics asBATCH are applied.
• MANUAL_IMMEDIATE - commit the offset immediately when the Acknowledgment.acknowledge() method is called by the listener.

The commitSync() or commitAsync() method on the consumer is used, depending on the syncCommits container property.

The Acknowledgment has this method:

public interface Acknowledgment {

    void acknowledge();

}

This gives the listener control over when offsets are committed.

@KafkaListener Annotation

The @KafkaListener annotation provides a mechanism for simple POJO listeners:

public class Listener {

    @KafkaListener(id = "foo", topics = "myTopic")
    public void listen(String data) {
        ...
    }

}

This mechanism requires an @EnableKafka annotation on one of your @Configuration classes and a listener container factory, which is used to configure the underlying ConcurrentMessageListenerContainer: by default, a bean with namekafkaListenerContainerFactory is expected.

@Configuration
@EnableKafka
public class KafkaConfig {

    @Bean
    KafkaListenerContainerFactory<ConcurrentMessageListenerContainer<Integer, String>>
                        kafkaListenerContainerFactory() {
        ConcurrentKafkaListenerContainerFactory<Integer, String> factory =
                                new ConcurrentKafkaListenerContainerFactory<>();
        factory.setConsumerFactory(consumerFactory());
        factory.setConcurrency(3);
        factory.getContainerProperties().setPollTimeout(3000);
        return factory;
    }

    @Bean
    public ConsumerFactory<Integer, String> consumerFactory() {
        return new DefaultKafkaConsumerFactory<>(consumerConfigs());
    }

    @Bean
    public Map<String, Object> consumerConfigs() {
        Map<String, Object> props = new HashMap<>();
        props.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, embeddedKafka.getBrokersAsString());
        ...
        return props;
    }
}

Notice that to set container properties, you must use the getContainerProperties() method on the factory. It is used as a template for the actual properties injected into the container.

You can also configure POJO listeners with explicit topics and partitions (and, optionally, their initial offsets):

@KafkaListener(id = "bar", topicPartitions =
        { @TopicPartition(topic = "topic1", partitions = { "0", "1" }),
          @TopicPartition(topic = "topic2", partitions = "0",
             partitionOffsets = @PartitionOffset(partition = "1", initialOffset = "100"))
        })
public void listen(ConsumerRecord<?, ?> record) {
    ...
}

Each partition can be specified in the partitions or partitionOffsets attribute, but not both.

When using manual AckMode, the listener can also be provided with the Acknowledgment; this example also shows how to use a different container factory.

@KafkaListener(id = "baz", topics = "myTopic",
          containerFactory = "kafkaManualAckListenerContainerFactory")
public void listen(String data, Acknowledgment ack) {
    ...
    ack.acknowledge();
}

Finally, metadata about the message is available from message headers:

@KafkaListener(id = "qux", topicPattern = "myTopic1")
public void listen(@Payload String foo,
        @Header(KafkaHeaders.RECEIVED_MESSAGE_KEY) Integer key,
        @Header(KafkaHeaders.RECEIVED_PARTITION_ID) int partition,
        @Header(KafkaHeaders.RECEIVED_TOPIC) String topic) {
    ...
}

Starting with version 1.1, @KafkaListener methods can be configured to receive the entire batch of consumer records received from the consumer poll. To configure the listener container factory to create batch listeners, set the batchListener property:

@Bean
public KafkaListenerContainerFactory<?> batchFactory() {
    ConcurrentKafkaListenerContainerFactory<Integer, String> factory =
            new ConcurrentKafkaListenerContainerFactory<>();
    factory.setConsumerFactory(consumerFactory());
    factory.setBatchListener(true);  // <<<<<<<<<<<<<<<<<<<<<<<<<
    return factory;
}

To receive a simple list of payloads:

@KafkaListener(id = "list", topics = "myTopic", containerFactory = "batchFactory")
public void listen(List<String> list) {
    ...
}

The topic, partition, offset etc are available in headers which parallel the payloads:

@KafkaListener(id = "list", topics = "myTopic", containerFactory = "batchFactory")
public void listen(List<String> list,
        @Header(KafkaHeaders.RECEIVED_MESSAGE_KEY) List<Integer> keys,
        @Header(KafkaHeaders.RECEIVED_PARTITION_ID) List<Integer> partitions,
        @Header(KafkaHeaders.RECEIVED_TOPIC) List<String> topics,
        @Header(KafkaHeaders.OFFSET) List<Long> offsets) {
    ...
}

Alternatively you can receive a List of Message<?> objects with each offset, etc in each message, but it must be the only parameter (aside from an optional Acknowledgment when using manual commits) defined on the method:

@KafkaListener(id = "listMsg", topics = "myTopic", containerFactory = "batchFactory")
public void listen14(List<Message<?>> list) {
    ...
}

@KafkaListener(id = "listMsgAck", topics = "myTopic", containerFactory = "batchFactory")
public void listen15(List<Message<?>> list, Acknowledgment ack) {
    ...
}

You can also receive a list of ConsumerRecord<?, ?> objects but it must be the only parameter (aside from an optionalAcknowledgment when using manual commits) defined on the method:

@KafkaListener(id = "listCRs", topics = "myTopic", containerFactory = "batchFactory")
public void listen(List<ConsumerRecord<Integer, String>> list) {
    ...
}

@KafkaListener(id = "listCRsAck", topics = "myTopic", containerFactory = "batchFactory")
public void listen(List<ConsumerRecord<Integer, String>> list, Acknowledgment ack) {
    ...
}

Filtering Messages

In certain scenarios, such as rebalancing, a message may be redelivered that has already been processed. The framework cannot know whether such a message has been processed or not, that is an application-level function. This is known as the Idempotent Receiver pattern and Spring Integration provides an implementation thereof.

The Spring for Apache Kafka project also provides some assistance by means of the FilteringMessageListenerAdapter class, which can wrap your MessageListener. This class takes an implementation of RecordFilterStrategy where you implement the filter method to signal that a message is a duplicate and should be discarded.

A FilteringAcknowledgingMessageListenerAdapter is also provided for wrapping an AcknowledgingMessageListener. This has an additional property ackDiscarded which indicates whether the adapter should acknowledge the discarded record; it istrue by default.

When using @KafkaListener, set the RecordFilterStrategy (and optionally ackDiscarded) on the container factory and the listener will be wrapped in the appropriate filtering adapter.

Finally, FilteringBatchMessageListenerAdapter and FilteringBatchAcknowledgingMessageListenerAdapter are provided, for when using a batch message listener.

Retrying Deliveries

If your listener throws an exception, the default behavior is to invoke the ErrorHandler, if configured, or logged otherwise.

To retry deliveries, convenient listener adapters - RetryingMessageListenerAdapter andRetryingAcknowledgingMessageListenerAdapter are provided, depending on whether you are using a MessageListener or an AcknowledgingMessageListener.

These can be configured with a RetryTemplate and RecoveryCallback<Void> - see the spring-retry project for information about these components. If a recovery callback is not provided, the exception is thrown to the container after retries are exhausted. In that case, the ErrorHandler will be invoked, if configured, or logged otherwise.

When using @KafkaListener, set the RetryTemplate (and optionally recoveryCallback) on the container factory and the listener will be wrapped in the appropriate retrying adapter.

A retry adapter is not provided for any of the batch message listeners.

Detecting Idle Asynchronous Consumers

While efficient, one problem with asynchronous consumers is detecting when they are idle - users might want to take some action if no messages arrive for some period of time.

You can configure the listener container to publish a ListenerContainerIdleEvent when some time passes with no message delivery. While the container is idle, an event will be published every idleEventInterval milliseconds.

To configure this feature, set the idleEventInterval on the container:

@Bean
public KafKaMessageListenerContainer(ConnectionFactory connectionFactory) {
    ContainerProperties containerProps = new ContainerProperties("topic1", "topic2");
    ...
    containerProps.setIdleEventInterval(60000L);
    ...
    KafKaMessageListenerContainer<String, String> container = new KafKaMessageListenerContainer<>(...);
    return container;
}

Or, for a @KafkaListener…

@Bean
public ConcurrentKafkaListenerContainerFactory kafkaListenerContainerFactory() {
    ConcurrentKafkaListenerContainerFactory<String, String> factory =
                new ConcurrentKafkaListenerContainerFactory<>();
    ...
    factory.getContainerProperties().setIdleEventInterval(60000L);
    ...
    return factory;
}

In each of these cases, an event will be published once per minute while the container is idle.

Event Consumption

You can capture these events by implementing ApplicationListener - either a general listener, or one narrowed to only receive this specific event. You can also use @EventListener, introduced in Spring Framework 4.2.

The following example combines the @KafkaListener and @EventListener into a single class. It’s important to understand that the application listener will get events for all containers so you may need to check the listener id if you want to take specific action based on which container is idle. You can also use the @EventListener condition for this purpose.

The events have 4 properties:

• source - the listener container instance
• id - the listener id (or container bean name)
• idleTime - the time the container had been idle when the event was published
• topicPartitions - the topics/partitions that the container was assigned at the time the event was generated

public class Listener { @KafkaListener(id = "qux", topics = "annotated") public void listen4(@Payload String foo, Acknowledgment ack) { ... } @EventListener(condition = "event.listenerId.startsWith('qux-')") public void eventHandler(ListenerContainerIdleEvent event) { this.event = event; eventLatch.countDown(); } }

Current Positions when Idle

Note that you can obtain the current positions when idle is detected by implementing ConsumerSeekAware in your listener; seeonIdleContainer() in `Seeking to a Specific Offset.

Topic/Partition Initial Offset

There are several ways to set the initial offset for a partition.

When manually assigning partitions, simply set the initial offset (if desired) in the configured TopicPartitionInitialOffsetarguments (see Message Listener Containers). You can also seek to a specific offset at any time.

When using group management where the broker assigns partitions:

• For a new group.id, the initial offset is determined by the auto.offset.reset consumer property (earliest orlatest).
• For an existing group id, the initial offset is the current offset for that group id. You can, however, seek to a specific offset during initialization (or at any time thereafter).

Seeking to a Specific Offset

In order to seek, your listener must implement ConsumerSeekAware which has the following methods:

void registerSeekCallback(ConsumerSeekCallback callback);

void onPartitionsAssigned(Map<TopicPartition, Long> assignments, ConsumerSeekCallback callback);

void onIdleContainer(Map<TopicPartition, Long> assignments, ConsumerSeekCallback callback);

The first is called when the container is started; this callback should be used when seeking at some arbitrary time after initialization. You should save a reference to the callback; if you are using the same listener in multiple containers (or in aConcurrentMessageListenerContainer) you should store the callback in a ThreadLocal or some other structure keyed by the listener Thread.

When using group management, the second method is called when assignments change. You can use this method, for example, for setting initial offsets for the partitions, by calling the callback; you must use the callback argument, not the one passed intoregisterSeekCallback. This method will never be called if you explicitly assign partitions yourself; use theTopicPartitionInitialOffset in that case.

The callback has one method:

void seek(String topic, int partition, long offset);

You can also perform seek operations from onIdleContainer() when an idle container is detected; see Detecting Idle Asynchronous Consumers for how to enable idle container detection.

To arbitrarily seek at runtime, use the callback reference from the registerSeekCallback for the appropriate thread.

Serialization/Deserialization and Message Conversion

Apache Kafka provides a high-level API for serializing/deserializing record values as well as their keys. It is present with theorg.apache.kafka.common.serialization.Serializer<T> and org.apache.kafka.common.serialization.Deserializer<T>abstractions with some built-in implementations. Meanwhile we can specify simple (de)serializer classes using Producer and/or Consumer configuration properties, e.g.:

props.put(ConsumerConfig.KEY_DESERIALIZER_CLASS_CONFIG, IntegerDeserializer.class);
props.put(ConsumerConfig.VALUE_DESERIALIZER_CLASS_CONFIG, StringDeserializer.class);
...
props.put(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG, IntegerSerializer.class);
props.put(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG, StringSerializer.class);

for more complex or particular cases, the KafkaConsumer, and therefore KafkaProducer, provides overloaded constructors to accept (De)Serializer instances for keys and/or values, respectively.

To meet this API, the DefaultKafkaProducerFactory and DefaultKafkaConsumerFactory also provide properties to allow to inject a custom (De)Serializer to target Producer/Consumer.

For this purpose, Spring for Apache Kafka also provides JsonSerializer/JsonDeserializer implementations based on the Jackson JSON object mapper. The JsonSerializer is quite simple and just allows writing any Java object as a JSON byte[], the JsonDeserializer requires an additional Class<?> targetType argument to allow the deserialization of a consumedbyte[] to the proper target object.

JsonDeserializer<Bar> barDeserializer = new JsonDeserializer<>(Bar.class);

Both JsonSerializer and JsonDeserializer can be customized with an ObjectMapper. You can also extend them to implement some particular configuration logic in the configure(Map<String, ?> configs, boolean isKey) method.

Although the Serializer/Deserializer API is quite simple and flexible from the low-level Kafka Consumer and Producerperspective, you might need more flexibility at the Spring Messaging level, either when using @KafkaListener or Spring Integration. To easily convert to/from org.springframework.messaging.Message, Spring for Apache Kafka provides aMessageConverter abstraction with the MessagingMessageConverter implementation and its StringJsonMessageConvertercustomization. The MessageConverter can be injected into KafkaTemplate instance directly and viaAbstractKafkaListenerContainerFactory bean definition for the @KafkaListener.containerFactory() property:

@Bean
public KafkaListenerContainerFactory<?> kafkaJsonListenerContainerFactory() {
    ConcurrentKafkaListenerContainerFactory<Integer, String> factory =
        new ConcurrentKafkaListenerContainerFactory<>();
    factory.setConsumerFactory(consumerFactory());
    factory.setMessageConverter(new StringJsonMessageConverter());
    return factory;
}
...
@KafkaListener(topics = "jsonData",
                containerFactory = "kafkaJsonListenerContainerFactory")
public void jsonListener(Foo foo) {
...
}

When using a @KafkaListener, the parameter type is provided to the message converter to assist with the conversion.

Log Compaction

When using Log Compaction, it is possible to send and receive messages with null payloads which identifies the deletion of a key.

Starting with version 1.0.3, this is now fully supported.

To send a null payload using the KafkaTemplate simply pass null into the value argument of the send() methods. One exception to this is the send(Message<?> message) variant. Since spring-messaging Message<?> cannot have a nullpayload, a special payload type KafkaNull is used and the framework will send null. For convenience, the staticKafkaNull.INSTANCE is provided.

When using a message listener container, the received ConsumerRecord will have a null value().

To configure the @KafkaListener to handle null payloads, you must use the @Payload annotation with required = false; you will usually also need the key so your application knows which key was "deleted":

@KafkaListener(id = "deletableListener", topics = "myTopic")
public void listen(@Payload(required = false) String value, @Header(KafkaHeaders.RECEIVED_MESSAGE_KEY) String key) {
    // value == null represents key deletion
}

When using a class-level @KafkaListener, some additional configuration is needed - a @KafkaHandler method with aKafkaNull payload:

@KafkaListener(id = "multi", topics = "myTopic")
static class MultiListenerBean {

    private final CountDownLatch latch1 = new CountDownLatch(2);

    @KafkaHandler
    public void listen(String foo) {
        ...
    }

    @KafkaHandler
    public void listen(Integer bar) {
        ...
    }

    @KafkaHandler
    public void delete(@Payload(required = false) KafkaNull nul, @Header(KafkaHeaders.RECEIVED_MESSAGE_KEY) int key) {
        ...
    }

}