As of Kafka 1.0.0, use KafkaPrincipalBuilder instead. This will be removed in a future major release.

This class has been deprecated and will be removed in a future release. Please use Deserializer instead.

This class has been deprecated and will be removed in a future release. Please use Serializer instead.

since 2.4 release; will be removed in 3.0.0 via KAFKA-7785

As of 1.0.0. This field will be removed in a future major release.

As of 1.0.0. This field will be removed in a future major release.

As of 1.0.0. This field will be removed in a future major release.

As of 1.0.0. This field will be removed in a future major release.

As of 1.0.0. This field will be removed in a future major release.

As of 1.0.0. This field will be removed in a future major release. In recent versions, the config is optional and there is no default.

As of 1.0.0. This field will be removed in a future major release.

As of 1.0.0. This field will be removed in a future major release.

As of 1.0.0. This field will be removed in a future major release.

As of 1.0.0. This field will be removed in a future major release.

since 2.7

since 2.7; use StreamsConfig.TOPOLOGY_OPTIMIZATION_CONFIG instead

Since 2.7.0. This will be removed in a future major release.

As of Kafka 0.11.0. Because of the potential for message format conversion on the broker, the checksum returned by the broker may not match what was computed by the producer. It is therefore unsafe to depend on this checksum for end-to-end delivery guarantees. Additionally, message format v2 does not include a record-level checksum (for performance, the record checksum was replaced with a batch checksum). To maintain compatibility, a partial checksum computed from the record timestamp, serialized key size, and serialized value size is returned instead, but this should not be depended on for end-to-end reliability.

Since 2.0. Use KafkaConsumer.close(Duration) or KafkaConsumer.close() .

since 2.4 Use KafkaConsumer.committed(Set) instead

Since 2.0. Use KafkaConsumer.poll(Duration) , which does not block beyond the timeout awaiting partition assignment. See KIP-266 for more information.

Use MockConsumer.setPollException(KafkaException) instead

please use NoOffsetForPartitionException.partitions

Since 2.7.0. This will be removed in a future major release.

As of Kafka 0.11.0. Because of the potential for message format conversion on the broker, the computed checksum may not match what was stored on the broker, or what will be returned to the consumer. It is therefore unsafe to depend on this checksum for end-to-end delivery guarantees. Additionally, message format v2 does not include a record-level checksum (for performance, the record checksum was replaced with a batch checksum). To maintain compatibility, a partial checksum computed from the record timestamp, serialized key size, and serialized value size is returned instead, but this should not be depended on for end-to-end reliability.

Use MessageFormatter.configure(Map) instead, this method is for backward compatibility with the older Formatter interface

As of 1.0.0, use Metric.metricValue() instead. This will be removed in a future major release.

As of 1.0.0. This method will be removed in a future major release.

Use SinkTask.open(Collection) for partition initialization.

Use SinkTask.close(Collection) instead for partition cleanup.

Use SourceTask.commitRecord(SourceRecord, RecordMetadata) instead.

Not called by Kafka Streams, which now uses KafkaClientSupplier.getAdmin(java.util.Map<java.lang.String, java.lang.Object>) instead.

Use KafkaStreams.close(Duration) instead; note, that KafkaStreams.close(Duration) has different semantics and does not block on zero, e.g., `Duration.ofMillis(0)`.

Since 2.5. Use KafkaStreams.queryMetadataForKey(String, Object, Serializer) instead.

Since 2.8.0. Use KafkaStreams.setUncaughtExceptionHandler(StreamsUncaughtExceptionHandler) instead.

since 2.5 release; use KafkaStreams.store(StoreQueryParameters) instead

Use KeyQueryMetadata.activeHost() instead.

Use KeyQueryMetadata.partition() instead.

Use KeyQueryMetadata.standbyHosts() instead.

this method will be removed in a in a future release

use Joined.as(String) instead

Use JoinWindows.after(Duration) instead

Use JoinWindows.before(Duration) instead.

since 2.1. This function should not be used anymore, since JoinWindows.until(long) is deprecated in favor of JoinWindows.grace(Duration) .

Use JoinWindows.of(Duration) instead.

since 2.1. Use JoinWindows.grace(Duration) instead.

since 2.8. Use KStream.split() instead.

since 2.1. Use KStream.groupBy(KeyValueMapper, Grouped) instead

since 2.1. Use KStream.groupByKey(Grouped) instead

since 2.4. Use KStream.join(KStream, ValueJoiner, JoinWindows, StreamJoined) instead.

since 2.4. Use KStream.leftJoin(KStream, ValueJoiner, JoinWindows, StreamJoined) instead.

since 2.4. Use KStream.outerJoin(KStream, ValueJoiner, JoinWindows, StreamJoined) instead.

since 2.6; use KStream.repartition() instead

since 2.1. Use KTable.groupBy(KeyValueMapper, Grouped) instead

since 2.1. Use Materialized.retention instead.

since 2.1. Use Materialized.retention or directly configure the retention in a store supplier and use Materialized.as(SessionBytesStoreSupplier) .

Use SessionWindows.with(Duration) instead.

Use TimeWindows.advanceBy(Duration) instead

since 2.1. Use Materialized.retention instead.

Use TimeWindows.of(Duration) instead

since 2.1. Use Materialized.retention or directly configure the retention in a store supplier and use Materialized.as(WindowBytesStoreSupplier) .

since 2.1. Use Materialized.retention instead.

Use UnlimitedWindows.startOn(Instant) instead

since 2.1.

since 2.1. Use Materialized.retention instead.

since 2.1 Override segmentInterval() instead.

since 2.1. Use Materialized.withRetention(Duration) or directly configure the retention in a store supplier and use Materialized.as(WindowBytesStoreSupplier) .

please use ProcessorContext.forward(Object, Object, To) instead

Use ProcessorContext.schedule(Duration, PunctuationType, Punctuator) instead

Since 2.7.0. Callers should invoke StateStore.init(StateStoreContext, StateStore) instead. Implementers may choose to implement this method for backward compatibility or to throw an informative exception instead.

Use ReadOnlyWindowStore.fetch(Object, Instant, Instant) instead

Use ReadOnlyWindowStore.fetchAll(Instant, Instant) instead

since 2.1 Use Stores.persistentSessionStore(String, Duration) instead

since 2.1 Use Stores.persistentWindowStore(String, Duration, Duration, boolean) instead

since 2.1. Use WindowBytesStoreSupplier.segmentIntervalMs() instead.

as timestamp is not provided for the key-value pair, this causes inconsistency to identify the window frame to which the key belongs. Use WindowStore.put(Object, Object, long) instead.

Use StreamsBuilder.addGlobalStore(StoreBuilder, String, Consumed, ProcessorSupplier) instead.

use StreamsConfig.getMainConsumerConfigs(String, String, int)

since 2.5. Use addLatencyRateTotalSensor() instead.

since 2.5. Use addRateTotalSensor() instead.

since 2.5. Use Sensor#record() instead.

since 2.5. Use Sensor#record() instead.

Since 2.7.0. Use Topology.addGlobalStore(StoreBuilder, String, Deserializer, Deserializer, String, String, ProcessorSupplier) instead.

Since 2.7.0 Use Topology.addProcessor(String, ProcessorSupplier, String...) instead.

use TopologyDescription.Source.topicSet() or TopologyDescription.Source.topicPattern() instead

Since 2.4 use TopologyTestDriver.advanceWallClockTime(Duration) instead

Since 2.4 use methods of TestInputTopic instead

Since 2.4 use methods of TestOutputTopic instead

Since 2.6.0. Use JmxReporter() Initialize JmxReporter with JmxReporter.contextChange(MetricsContext) Populate prefix by adding _namespace/prefix key value pair to MetricsContext

use KafkaStreams(Topology, Properties) instead

Since 2.4 use TopologyTestDriver(Topology, Properties, Instant)