Class: Kafka::Client

Inherits:
Object
  • Object
show all
Defined in:
lib/kafka/client.rb

Instance Method Summary collapse

Constructor Details

#initialize(seed_brokers:, client_id: "ruby-kafka", logger: nil, connect_timeout: nil, socket_timeout: nil, ssl_ca_cert_file_path: nil, ssl_ca_cert: nil, ssl_client_cert: nil, ssl_client_cert_key: nil, sasl_gssapi_principal: nil, sasl_gssapi_keytab: nil, sasl_plain_authzid: '', sasl_plain_username: nil, sasl_plain_password: nil, sasl_scram_username: nil, sasl_scram_password: nil, sasl_scram_mechanism: nil, ssl_ca_certs_from_system: false) ⇒ Client

Initializes a new Kafka client.



58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
 
# File 'lib/kafka/client.rb', line 58

def initialize(seed_brokers:, client_id: "ruby-kafka", logger: nil, connect_timeout: nil, socket_timeout: nil,
               ssl_ca_cert_file_path: nil, ssl_ca_cert: nil, ssl_client_cert: nil, ssl_client_cert_key: nil,
               sasl_gssapi_principal: nil, sasl_gssapi_keytab: nil,
               sasl_plain_authzid: '', sasl_plain_username: nil, sasl_plain_password: nil,
               sasl_scram_username: nil, sasl_scram_password: nil, sasl_scram_mechanism: nil, ssl_ca_certs_from_system: false)
  @logger = logger || Logger.new(nil)
  @instrumenter = Instrumenter.new(client_id: client_id)
  @seed_brokers = normalize_seed_brokers(seed_brokers)

  ssl_context = SslContext.build(
    ca_cert_file_path: ssl_ca_cert_file_path,
    ca_cert: ssl_ca_cert,
    client_cert: ssl_client_cert,
    client_cert_key: ssl_client_cert_key,
    ca_certs_from_system: ssl_ca_certs_from_system,
  )

  sasl_authenticator = SaslAuthenticator.new(
    sasl_gssapi_principal: sasl_gssapi_principal,
    sasl_gssapi_keytab: sasl_gssapi_keytab,
    sasl_plain_authzid: sasl_plain_authzid,
    sasl_plain_username: sasl_plain_username,
    sasl_plain_password: sasl_plain_password,
    sasl_scram_username: sasl_scram_username,
    sasl_scram_password: sasl_scram_password,
    sasl_scram_mechanism: sasl_scram_mechanism,
    logger: @logger
  )

  if sasl_authenticator.enabled? && ssl_context.nil?
    raise ArgumentError, "SASL authentication requires that SSL is configured"
  end

  @connection_builder = ConnectionBuilder.new(
    client_id: client_id,
    connect_timeout: connect_timeout,
    socket_timeout: socket_timeout,
    ssl_context: ssl_context,
    logger: @logger,
    instrumenter: @instrumenter,
    sasl_authenticator: sasl_authenticator
  )

  @cluster = initialize_cluster
end

Instance Method Details

#alter_topic(name, configs = {}) ⇒ nil

Note:

This is an alpha level API and is subject to change.

Alter the configuration of a topic.

Configuration keys must match Kafka's topic-level configs.

Examples:

Describing the cleanup policy config of a topic

kafka = Kafka.new(["kafka1:9092"])
kafka.alter_topic("my-topic", "cleanup.policy" => "delete", "max.message.byte" => "100000")



539
540
541
 
# File 'lib/kafka/client.rb', line 539

def alter_topic(name, configs = {})
  @cluster.alter_topic(name, configs)
end

#apis ⇒ Object 



624
625
626
 
# File 'lib/kafka/client.rb', line 624

def apis
  @cluster.apis
end

#async_producer(delivery_interval: 0, delivery_threshold: 0, max_queue_size: 1000, **options) ⇒ AsyncProducer

Creates a new AsyncProducer instance.

All parameters allowed by #producer can be passed. In addition to this, a few extra parameters can be passed when creating an async producer.

See Also:



253
254
255
256
257
258
259
260
261
262
263
264
 
# File 'lib/kafka/client.rb', line 253

def async_producer(delivery_interval: 0, delivery_threshold: 0, max_queue_size: 1000, **options)
  sync_producer = producer(**options)

  AsyncProducer.new(
    sync_producer: sync_producer,
    delivery_interval: delivery_interval,
    delivery_threshold: delivery_threshold,
    max_queue_size: max_queue_size,
    instrumenter: @instrumenter,
    logger: @logger,
  )
end

#close ⇒ nil

Closes all connections to the Kafka brokers and frees up used resources.



631
632
633
 
# File 'lib/kafka/client.rb', line 631

def close
  @cluster.disconnect
end

#consumer(group_id:, session_timeout: 30, offset_commit_interval: 10, offset_commit_threshold: 0, heartbeat_interval: 10, offset_retention_time: nil) ⇒ Consumer

Creates a new Kafka consumer.



281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
 
# File 'lib/kafka/client.rb', line 281

def consumer(group_id:, session_timeout: 30, offset_commit_interval: 10, offset_commit_threshold: 0, heartbeat_interval: 10, offset_retention_time: nil)
  cluster = initialize_cluster

  instrumenter = DecoratingInstrumenter.new(@instrumenter, {
    group_id: group_id,
  })

  # The Kafka protocol expects the retention time to be in ms.
  retention_time = (offset_retention_time && offset_retention_time * 1_000) || -1

  group = ConsumerGroup.new(
    cluster: cluster,
    logger: @logger,
    group_id: group_id,
    session_timeout: session_timeout,
    retention_time: retention_time,
    instrumenter: instrumenter,
  )

  fetcher = Fetcher.new(
    cluster: initialize_cluster,
    logger: @logger,
    instrumenter: instrumenter,
  )

  offset_manager = OffsetManager.new(
    cluster: cluster,
    group: group,
    fetcher: fetcher,
    logger: @logger,
    commit_interval: offset_commit_interval,
    commit_threshold: offset_commit_threshold,
    offset_retention_time: offset_retention_time
  )

  heartbeat = Heartbeat.new(
    group: group,
    interval: heartbeat_interval,
  )

  Consumer.new(
    cluster: cluster,
    logger: @logger,
    instrumenter: instrumenter,
    group: group,
    offset_manager: offset_manager,
    fetcher: fetcher,
    session_timeout: session_timeout,
    heartbeat: heartbeat,
  )
end

#create_partitions_for(name, num_partitions: 1, timeout: 30) ⇒ nil

Create partitions for a topic.

the topic partitions to be added.



551
552
553
 
# File 'lib/kafka/client.rb', line 551

def create_partitions_for(name, num_partitions: 1, timeout: 30)
  @cluster.create_partitions_for(name, num_partitions: num_partitions, timeout: timeout)
end

#create_topic(name, num_partitions: 1, replication_factor: 1, timeout: 30, config: {}) ⇒ nil

Creates a topic in the cluster.

Examples:

Creating a topic with log compaction

# Enable log compaction:
config = { "cleanup.policy" => "compact" }

# Create the topic:
kafka.create_topic("dns-mappings", config: config)

Raises:



486
487
488
489
490
491
492
493
494
 
# File 'lib/kafka/client.rb', line 486

def create_topic(name, num_partitions: 1, replication_factor: 1, timeout: 30, config: {})
  @cluster.create_topic(
    name,
    num_partitions: num_partitions,
    replication_factor: replication_factor,
    timeout: timeout,
    config: config,
  )
end

#delete_topic(name, timeout: 30) ⇒ nil

Delete a topic in the cluster.



502
503
504
 
# File 'lib/kafka/client.rb', line 502

def delete_topic(name, timeout: 30)
  @cluster.delete_topic(name, timeout: timeout)
end

#deliver_message(value, key: nil, topic:, partition: nil, partition_key: nil, retries: 1) ⇒ nil

Delivers a single message to the Kafka cluster.

Note: Only use this API for low-throughput scenarios. If you want to deliver many messages at a high rate, or if you want to configure the way messages are sent, use the #producer or #async_producer APIs instead.



121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
 
# File 'lib/kafka/client.rb', line 121

def deliver_message(value, key: nil, topic:, partition: nil, partition_key: nil, retries: 1)
  create_time = Time.now

  message = PendingMessage.new(
    value,
    key,
    topic,
    partition,
    partition_key,
    create_time,
  )

  if partition.nil?
    partition_count = @cluster.partitions_for(topic).count
    partition = Partitioner.partition_for_key(partition_count, message)
  end

  buffer = MessageBuffer.new

  buffer.write(
    value: message.value,
    key: message.key,
    topic: message.topic,
    partition: partition,
    create_time: message.create_time,
  )

  @cluster.add_target_topics([topic])

  compressor = Compressor.new(
    instrumenter: @instrumenter,
  )

  operation = ProduceOperation.new(
    cluster: @cluster,
    buffer: buffer,
    required_acks: 1,
    ack_timeout: 10,
    compressor: compressor,
    logger: @logger,
    instrumenter: @instrumenter,
  )

  attempt = 1

  begin
    operation.execute

    unless buffer.empty?
      raise DeliveryFailed.new(nil, [message])
    end
  rescue Kafka::Error => e
    @cluster.mark_as_stale!

    if attempt >= (retries + 1)
      raise
    else
      attempt += 1
      @logger.warn "Error while delivering message, #{e.class}: #{e.message}; retrying after 1s..."

      sleep 1

      retry
    end
  end
end

#describe_topic(name, configs = []) ⇒ Hash<String, String>

Note:

This is an alpha level API and is subject to change.

Describe the configuration of a topic.

Retrieves the topic configuration from the Kafka brokers. Configuration names refer to Kafka's topic-level configs.

Examples:

Describing the cleanup policy config of a topic

kafka = Kafka.new(["kafka1:9092"])
kafka.describe_topic("my-topic", ["cleanup.policy"])
#=> { "cleanup.policy" => "delete" }



521
522
523
 
# File 'lib/kafka/client.rb', line 521

def describe_topic(name, configs = [])
  @cluster.describe_topic(name, configs)
end

#each_message(topic:, start_from_beginning: true, max_wait_time: 5, min_bytes: 1, max_bytes: 1048576, &block) ⇒ nil

Enumerate all messages in a topic.



440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
 
# File 'lib/kafka/client.rb', line 440

def each_message(topic:, start_from_beginning: true, max_wait_time: 5, min_bytes: 1, max_bytes: 1048576, &block)
  default_offset ||= start_from_beginning ? :earliest : :latest
  offsets = Hash.new { default_offset }

  loop do
    operation = FetchOperation.new(
      cluster: @cluster,
      logger: @logger,
      min_bytes: min_bytes,
      max_wait_time: max_wait_time,
    )

    @cluster.partitions_for(topic).map(&:partition_id).each do |partition|
      partition_offset = offsets[partition]
      operation.fetch_from_partition(topic, partition, offset: partition_offset, max_bytes: max_bytes)
    end

    batches = operation.execute

    batches.each do |batch|
      batch.messages.each(&block)
      offsets[batch.partition] = batch.last_offset + 1
    end
  end
end

#fetch_messages(topic:, partition:, offset: :latest, max_wait_time: 5, min_bytes: 1, max_bytes: 1048576, retries: 1) ⇒ Array<Kafka::FetchedMessage>

Fetches a batch of messages from a single partition. Note that it's possible to get back empty batches.

The starting point for the fetch can be configured with the :offset argument. If you pass a number, the fetch will start at that offset. However, there are two special Symbol values that can be passed instead:

  • :earliest — the first offset in the partition.
  • :latest — the next offset that will be written to, effectively making the call block until there is a new message in the partition.

The Kafka protocol specifies the numeric values of these two options: -2 and -1, respectively. You can also pass in these numbers directly.

Example

When enumerating the messages in a partition, you typically fetch batches sequentially.

offset = :earliest

loop do
  messages = kafka.fetch_messages(
    topic: "my-topic",
    partition: 42,
    offset: offset,
  )

  messages.each do |message|
    puts message.offset, message.key, message.value

    # Set the next offset that should be read to be the subsequent
    # offset.
    offset = message.offset + 1
  end
end

See a working example in examples/simple-consumer.rb.



392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
 
# File 'lib/kafka/client.rb', line 392

def fetch_messages(topic:, partition:, offset: :latest, max_wait_time: 5, min_bytes: 1, max_bytes: 1048576, retries: 1)
  operation = FetchOperation.new(
    cluster: @cluster,
    logger: @logger,
    min_bytes: min_bytes,
    max_bytes: max_bytes,
    max_wait_time: max_wait_time,
  )

  operation.fetch_from_partition(topic, partition, offset: offset, max_bytes: max_bytes)

  attempt = 1

  begin
    operation.execute.flat_map {|batch| batch.messages }
  rescue Kafka::Error => e
    @cluster.mark_as_stale!

    if attempt >= (retries + 1)
      raise
    else
      attempt += 1
      @logger.warn "Error while fetching messages, #{e.class}: #{e.message}; retrying..."
      retry
    end
  end
end

#groups ⇒ Array<String>

Lists all consumer groups in the cluster



565
566
567
 
# File 'lib/kafka/client.rb', line 565

def groups
  @cluster.list_groups
end

#has_topic?(topic) ⇒ Boolean 



569
570
571
572
573
 
# File 'lib/kafka/client.rb', line 569

def has_topic?(topic)
  @cluster.clear_target_topics
  @cluster.add_target_topics([topic])
  @cluster.topics.include?(topic)
end

#last_offset_for(topic, partition) ⇒ Integer

Retrieve the offset of the last message in a partition. If there are no messages in the partition -1 is returned.



590
591
592
593
594
 
# File 'lib/kafka/client.rb', line 590

def last_offset_for(topic, partition)
  # The offset resolution API will return the offset of the "next" message to
  # be written when resolving the "latest" offset, so we subtract one.
  @cluster.resolve_offset(topic, partition, :latest) - 1
end

#last_offsets_for(*topics) ⇒ Hash<String, Hash<Integer, Integer>>

Retrieve the offset of the last message in each partition of the specified topics.

Examples:

last_offsets_for('topic-1', 'topic-2') # =>
# {
#   'topic-1' => { 0 => 100, 1 => 100 },
#   'topic-2' => { 0 => 100, 1 => 100 }
# }



606
607
608
609
610
611
612
613
 
# File 'lib/kafka/client.rb', line 606

def last_offsets_for(*topics)
  @cluster.add_target_topics(topics)
  topics.map {|topic|
    partition_ids = @cluster.partitions_for(topic).collect(&:partition_id)
    partition_offsets = @cluster.resolve_offsets(topic, partition_ids, :latest)
    [topic, partition_offsets.collect { |k, v| [k, v - 1] }.to_h]
  }.to_h
end

#partitions_for(topic) ⇒ Integer

Counts the number of partitions in a topic.



579
580
581
 
# File 'lib/kafka/client.rb', line 579

def partitions_for(topic)
  @cluster.partitions_for(topic).count
end

#producer(compression_codec: nil, compression_threshold: 1, ack_timeout: 5, required_acks: :all, max_retries: 2, retry_backoff: 1, max_buffer_size: 1000, max_buffer_bytesize: 10_000_000) ⇒ Kafka::Producer

Initializes a new Kafka producer.



218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
 
# File 'lib/kafka/client.rb', line 218

def producer(compression_codec: nil, compression_threshold: 1, ack_timeout: 5, required_acks: :all, max_retries: 2, retry_backoff: 1, max_buffer_size: 1000, max_buffer_bytesize: 10_000_000)
  compressor = Compressor.new(
    codec_name: compression_codec,
    threshold: compression_threshold,
    instrumenter: @instrumenter,
  )

  Producer.new(
    cluster: initialize_cluster,
    logger: @logger,
    instrumenter: @instrumenter,
    compressor: compressor,
    ack_timeout: ack_timeout,
    required_acks: required_acks,
    max_retries: max_retries,
    retry_backoff: retry_backoff,
    max_buffer_size: max_buffer_size,
    max_buffer_bytesize: max_buffer_bytesize,
  )
end

#supports_api?(api_key, version = nil) ⇒ Boolean

Check whether current cluster supports a specific version or not



620
621
622
 
# File 'lib/kafka/client.rb', line 620

def supports_api?(api_key, version = nil)
  @cluster.supports_api?(api_key, version)
end

#topics ⇒ Array<String>

Lists all topics in the cluster.



558
559
560
 
# File 'lib/kafka/client.rb', line 558

def topics
  @cluster.list_topics
end