Class: Kafka::Client

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

Constant Summary collapse

URI_SCHEMES = 
["kafka", "kafka+ssl"]

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) ⇒ Client

Initializes a new Kafka client.

Parameters:

  • seed_brokers (Array<String>, String) —

    the list of brokers used to initialize the client. Either an Array of connections, or a comma separated string of connections. A connection can either be a string of "host:port" or a full URI with a scheme. If there's a scheme it's ignored and only host/port are used.

  • client_id (String) (defaults to: "ruby-kafka") —

    the identifier for this application.

  • logger (Logger) (defaults to: nil) —

    the logger that should be used by the client.

  • connect_timeout (Integer, nil) (defaults to: nil) —

    the timeout setting for connecting to brokers. See BrokerPool#initialize.

  • socket_timeout (Integer, nil) (defaults to: nil) —

    the timeout setting for socket connections. See BrokerPool#initialize.

  • ssl_ca_cert (String, Array<String>, nil) (defaults to: nil) —

    a PEM encoded CA cert, or an Array of PEM encoded CA certs, to use with an SSL connection.

  • ssl_ca_cert_file_path (String, nil) (defaults to: nil) —

    a path on the filesystem to a PEM encoded CA cert to use with an SSL connection.

  • ssl_client_cert (String, nil) (defaults to: nil) —

    a PEM encoded client cert to use with an SSL connection. Must be used in combination with ssl_client_cert_key.

  • ssl_client_cert_key (String, nil) (defaults to: nil) —

    a PEM encoded client cert key to use with an SSL connection. Must be used in combination with ssl_client_cert.

  • sasl_gssapi_principal (String, nil) (defaults to: nil) —

    a KRB5 principal

  • sasl_gssapi_keytab (String, nil) (defaults to: nil) —

    a KRB5 keytab filepath

  • sasl_scram_username (String, nil) (defaults to: nil) —

    SCRAM username

  • sasl_scram_password (String, nil) (defaults to: nil) —

    SCRAM password

  • sasl_scram_mechanism (String, nil) (defaults to: nil) —

    Scram mechanism, either "sha256" or "sha512"



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
 
# File 'lib/kafka/client.rb', line 59

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)
  @logger = logger || Logger.new(nil)
  @instrumenter = Instrumenter.new(client_id: client_id)
  @seed_brokers = normalize_seed_brokers(seed_brokers)

  ssl_context = build_ssl_context(ssl_ca_cert_file_path, ssl_ca_cert, ssl_client_cert, ssl_client_cert_key)

  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
  )

  @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

#apis ⇒ Object 



512
513
514
 
# File 'lib/kafka/client.rb', line 512

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.

Parameters:

  • max_queue_size (Integer) (defaults to: 1000) —

    the maximum number of messages allowed in the queue.

  • delivery_threshold (Integer) (defaults to: 0) —

    if greater than zero, the number of buffered messages that will automatically trigger a delivery.

  • delivery_interval (Integer) (defaults to: 0) —

    if greater than zero, the number of seconds between automatic message deliveries.

Returns:

See Also:



244
245
246
247
248
249
250
251
252
253
254
255
 
# File 'lib/kafka/client.rb', line 244

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.

Returns:

  • (nil) 


519
520
521
 
# File 'lib/kafka/client.rb', line 519

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.

Parameters:

  • group_id (String) —

    the id of the group that the consumer should join.

  • session_timeout (Integer) (defaults to: 30) —

    the number of seconds after which, if a client hasn't contacted the Kafka cluster, it will be kicked out of the group.

  • offset_commit_interval (Integer) (defaults to: 10) —

    the interval between offset commits, in seconds.

  • offset_commit_threshold (Integer) (defaults to: 0) —

    the number of messages that can be processed before their offsets are committed. If zero, offset commits are not triggered by message processing.

  • heartbeat_interval (Integer) (defaults to: 10) —

    the interval between heartbeats; must be less than the session window.

  • offset_retention_time (Integer) (defaults to: nil) —

    the time period that committed offsets will be retained, in seconds. Defaults to the broker setting.

Returns:



272
273
274
275
276
277
278
279
280
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
 
# File 'lib/kafka/client.rb', line 272

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,
  )

  offset_manager = OffsetManager.new(
    cluster: cluster,
    group: group,
    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,
    session_timeout: session_timeout,
    heartbeat: heartbeat,
  )
end

#create_topic(name, **options) ⇒ Object 



454
455
456
 
# File 'lib/kafka/client.rb', line 454

def create_topic(name, **options)
  @cluster.create_topic(name, **options)
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.

Parameters:

  • value (String, nil) —

    the message value.

  • key (String, nil) (defaults to: nil) —

    the message key.

  • topic (String) —

    the topic that the message should be written to.

  • partition (Integer, nil) (defaults to: nil) —

    the partition that the message should be written to, or nil if either partition_key is passed or the partition should be chosen at random.

  • partition_key (String) (defaults to: nil) —

    a value used to deterministically choose a partition to write to.

  • retries (Integer) (defaults to: 1) —

    the number of times to retry the delivery before giving up.

Returns:

  • (nil) 


112
113
114
115
116
117
118
119
120
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
 
# File 'lib/kafka/client.rb', line 112

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

#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.

Parameters:

  • topic (String) —

    the topic to consume messages from.

  • start_from_beginning (Boolean) (defaults to: true) —

    whether to start from the beginning of the topic or just subscribe to new messages being produced. This only applies when first consuming a topic partition – once the consumer has checkpointed its progress, it will always resume from the last checkpoint.

  • max_wait_time (Integer) (defaults to: 5) —

    the maximum amount of time to wait before the server responds, in seconds.

  • min_bytes (Integer) (defaults to: 1) —

    the minimum number of bytes to wait for. If set to zero, the broker will respond immediately, but the response may be empty. The default is 1 byte, which means that the broker will respond as soon as a message is written to the partition.

  • max_bytes (Integer) (defaults to: 1048576) —

    the maximum number of bytes to include in the response message set. Default is 1 MB. You need to set this higher if you expect messages to be larger than this.

Returns:

  • (nil) 


428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
 
# File 'lib/kafka/client.rb', line 428

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>

Note:

This API is still alpha level. Don't try to use it in production.

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.

Parameters:

  • topic (String) —

    the topic that messages should be fetched from.

  • partition (Integer) —

    the partition that messages should be fetched from.

  • offset (Integer, Symbol) (defaults to: :latest) —

    the offset to start reading from. Default is the latest offset.

  • max_wait_time (Integer) (defaults to: 5) —

    the maximum amount of time to wait before the server responds, in seconds.

  • min_bytes (Integer) (defaults to: 1) —

    the minimum number of bytes to wait for. If set to zero, the broker will respond immediately, but the response may be empty. The default is 1 byte, which means that the broker will respond as soon as a message is written to the partition.

  • max_bytes (Integer) (defaults to: 1048576) —

    the maximum number of bytes to include in the response message set. Default is 1 MB. You need to set this higher if you expect messages to be larger than this.

Returns:



377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
 
# File 'lib/kafka/client.rb', line 377

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

#has_topic?(topic) ⇒ Boolean

Returns:

  • (Boolean) 


466
467
468
469
470
 
# File 'lib/kafka/client.rb', line 466

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.

Parameters:

  • topic (String)
  • partition (Integer)

Returns:

  • (Integer) —

    the offset of the last message in the partition, or -1 if there are no messages in the partition.



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

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 }
# }

Parameters:

  • topics (Array<String>) —

    topic names.

Returns:

  • (Hash<String, Hash<Integer, Integer>>) 


503
504
505
506
507
508
509
510
 
# File 'lib/kafka/client.rb', line 503

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.

Parameters:

  • topic (String)

Returns:

  • (Integer) —

    the number of partitions in the topic.



476
477
478
 
# File 'lib/kafka/client.rb', line 476

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.

Parameters:

  • ack_timeout (Integer) (defaults to: 5) —

    The number of seconds a broker can wait for replicas to acknowledge a write before responding with a timeout.

  • required_acks (Integer, Symbol) (defaults to: :all) —

    The number of replicas that must acknowledge a write, or :all if all in-sync replicas must acknowledge.

  • max_retries (Integer) (defaults to: 2) —

    the number of retries that should be attempted before giving up sending messages to the cluster. Does not include the original attempt.

  • retry_backoff (Integer) (defaults to: 1) —

    the number of seconds to wait between retries.

  • max_buffer_size (Integer) (defaults to: 1000) —

    the number of messages allowed in the buffer before new writes will raise BufferOverflow exceptions.

  • max_buffer_bytesize (Integer) (defaults to: 10_000_000) —

    the maximum size of the buffer in bytes. attempting to produce messages when the buffer reaches this size will result in BufferOverflow being raised.

  • compression_codec (Symbol, nil) (defaults to: nil) —

    the name of the compression codec to use, or nil if no compression should be performed. Valid codecs: :snappy and :gzip.

  • compression_threshold (Integer) (defaults to: 1) —

    the number of messages that needs to be in a message set before it should be compressed. Note that message sets are per-partition rather than per-topic or per-producer.

Returns:



209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
 
# File 'lib/kafka/client.rb', line 209

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

#topics ⇒ Array<String>

Lists all topics in the cluster.

Returns:

  • (Array<String>) —

    the list of topic names.



461
462
463
464
 
# File 'lib/kafka/client.rb', line 461

def topics
  @cluster.clear_target_topics
  @cluster.topics
end