Class: Fluent::Rdkafka2Output

Inherits:
Output
  • Object
show all
Includes:
KafkaPluginUtil::SSLSettings, KafkaPluginUtil::SaslSettings
Defined in:
lib/fluent/plugin/out_rdkafka2.rb

Defined Under Namespace

Classes: EnqueueRate

Constant Summary

Constants included from KafkaPluginUtil::SSLSettings

KafkaPluginUtil::SSLSettings::DummyFormatter

Instance Method Summary collapse

Methods included from KafkaPluginUtil::SaslSettings

included

Methods included from KafkaPluginUtil::SSLSettings

included, #pickup_ssl_endpoint, #read_ssl_file

Constructor Details

#initialize ⇒ Rdkafka2Output

Returns a new instance of Rdkafka2Output.



168
169
170
171
172
173
174
175
176
177
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 168

def initialize
  super

  @producers = nil
  @producers_mutex = nil
  @shared_producer = nil
  @enqueue_rate = nil
  @writing_threads_mutex = Mutex.new
  @writing_threads = Set.new
end

Instance Method Details

#build_config ⇒ Object 



250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
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
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 250

def build_config
  config = {:"bootstrap.servers" => @brokers}

  if @ssl_ca_cert && @ssl_ca_cert[0]
    ssl = true
    config[:"ssl.ca.location"] = @ssl_ca_cert[0]
    config[:"ssl.certificate.location"] = @ssl_client_cert if @ssl_client_cert
    config[:"ssl.key.location"] = @ssl_client_cert_key if @ssl_client_cert_key
    config[:"ssl.key.password"] = @ssl_client_cert_key_password if @ssl_client_cert_key_password
  end

  if @principal
    sasl = true
    config[:"sasl.mechanisms"] = "GSSAPI"
    config[:"sasl.kerberos.principal"] = @principal
    config[:"sasl.kerberos.service.name"] = @service_name if @service_name
    config[:"sasl.kerberos.keytab"] = @keytab if @keytab
  end

  if ssl && sasl
    security_protocol = "SASL_SSL"
  elsif ssl && !sasl
    security_protocol = "SSL"
  elsif !ssl && sasl
    security_protocol = "SASL_PLAINTEXT"
  else
    security_protocol = "PLAINTEXT"
  end
  config[:"security.protocol"] = security_protocol

  config[:"compression.codec"] = @compression_codec if @compression_codec
  config[:"message.send.max.retries"] = @max_send_retries if @max_send_retries
  config[:"request.required.acks"] = @required_acks if @required_acks
  config[:"request.timeout.ms"] = @ack_timeout * 1000 if @ack_timeout
  config[:"queue.buffering.max.ms"] = @rdkafka_buffering_max_ms if @rdkafka_buffering_max_ms
  config[:"queue.buffering.max.messages"] = @rdkafka_buffering_max_messages if @rdkafka_buffering_max_messages
  config[:"message.max.bytes"] = @rdkafka_message_max_bytes if @rdkafka_message_max_bytes
  config[:"batch.num.messages"] = @rdkafka_message_max_num if @rdkafka_message_max_num
  config[:"sasl.username"] = @username if @username
  config[:"sasl.password"] = @password if @password
  config[:"enable.idempotence"] = @idempotent if @idempotent

  @rdkafka_options.each { |k, v|
    config[k.to_sym] = v
  }

  config
end

#close_producer(producer) ⇒ Object 



349
350
351
352
353
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 349

def close_producer(producer)
  unless producer.close(10)
    log.warn("Queue is forcefully closed after 10 seconds wait")
  end
end

#configure(conf) ⇒ Object 



179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 179

def configure(conf)
  super
  log.instance_eval {
    def add(level, message = nil)
      if message.nil?
        if block_given?
          message = yield
        else
          return
        end
      end

      # Follow rdkakfa's log level. See also rdkafka-ruby's bindings.rb: https://github.com/appsignal/rdkafka-ruby/blob/e5c7261e3f2637554a5c12b924be297d7dca1328/lib/rdkafka/bindings.rb#L117
      case level
      when Logger::FATAL
        self.fatal(message)
      when Logger::ERROR
        self.error(message)
      when Logger::WARN
        self.warn(message)
      when Logger::INFO
        self.info(message)
      when Logger::DEBUG
        self.debug(message)
      else
        self.trace(message)
      end
    end
  }
  Rdkafka::Config.logger = log
  config = build_config
  @rdkafka = Rdkafka::Config.new(config)

  if @default_topic.nil?
    if @use_default_for_unknown_topic || @use_default_for_unknown_partition_error
      raise Fluent::ConfigError, "default_topic must be set when use_default_for_unknown_topic or use_default_for_unknown_partition_error is true"
    end
    if @chunk_keys.include?(@topic_key) && !@chunk_key_tag
      log.warn "Use '#{@topic_key}' field of event record for topic but no fallback. Recommend to set default_topic or set 'tag' in buffer chunk keys like <buffer #{@topic_key},tag>"
    end
  else
    if @chunk_key_tag
      log.warn "default_topic is set. Fluentd's event tag is not used for topic"
    end
  end

  formatter_conf = conf.elements('format').first
  unless formatter_conf
    raise Fluent::ConfigError, "<format> section is required."
  end
  unless formatter_conf["@type"]
    raise Fluent::ConfigError, "format/@type is required."
  end
  @formatter_proc = setup_formatter(formatter_conf)
  @topic_key_sym = @topic_key.to_sym

  @headers_from_record_accessors = {}
  @headers_from_record.each do |key, value|
    @headers_from_record_accessors[key] = record_accessor_create(value)
  end

  @exclude_field_accessors = @exclude_fields.map do |field|
    record_accessor_create(field)
  end

  @enqueue_rate = EnqueueRate.new(@max_enqueue_bytes_per_second) unless @max_enqueue_bytes_per_second.nil?

  @record_field_accessor = nil
  @record_field_accessor = record_accessor_create(@record_key) unless @record_key.nil?
end

#enqueue_with_retry(producer, topic, record_buf, message_key, partition, headers, time) ⇒ Object 



452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 452

def enqueue_with_retry(producer, topic, record_buf, message_key, partition, headers, time)
  attempt = 0
  actual_topic = topic

  loop do
    begin
      @enqueue_rate.raise_if_limit_exceeded(record_buf.bytesize) if @enqueue_rate
      return producer.produce(topic: actual_topic, payload: record_buf, key: message_key, partition: partition, headers: headers, timestamp: @use_event_time ? Time.at(time) : nil)
    rescue EnqueueRate::LimitExceeded => e
      @enqueue_rate.revert if @enqueue_rate
      duration = e.next_retry_clock - Fluent::Clock.now
      sleep(duration) if duration > 0.0
    rescue Exception => e
      @enqueue_rate.revert if @enqueue_rate

      if !e.respond_to?(:code)
        raise e
      end

      case e.code
      when :queue_full
        if attempt <= @max_enqueue_retries
          log.warn "Failed to enqueue message; attempting retry #{attempt} of #{@max_enqueue_retries} after #{@enqueue_retry_backoff}s"
          sleep @enqueue_retry_backoff
          attempt += 1
        else
          raise "Failed to enqueue message although tried retry #{@max_enqueue_retries} times"
        end
      # https://github.com/confluentinc/librdkafka/blob/c282ba2423b2694052393c8edb0399a5ef471b3f/src/rdkafka.h#LL309C9-L309C41
      # RD_KAFKA_RESP_ERR__UNKNOWN_TOPIC
      when :unknown_topic
        if @use_default_for_unknown_topic && actual_topic != @default_topic
          log.debug "'#{actual_topic}' topic not found. Retry with '#{@default_topic}' topic"
          actual_topic = @default_topic
          retry
        end
        raise e
      # https://github.com/confluentinc/librdkafka/blob/c282ba2423b2694052393c8edb0399a5ef471b3f/src/rdkafka.h#L305
      # RD_KAFKA_RESP_ERR__UNKNOWN_PARTITION
      when :unknown_partition
        if @use_default_for_unknown_partition_error && actual_topic != @default_topic
          log.debug "failed writing to topic '#{actual_topic}' with error '#{e.to_s}'. Writing message to topic '#{@default_topic}'"
          actual_topic = @default_topic
          retry
        end

        raise e
      else
        if unrecoverable_error_codes.include?(e.code.to_s)
          # some of the errors should be handled as an unrecoverable error
          raise Fluent::UnrecoverableError, "Rejected due to #{e}"
        else
          raise e
        end
      end
    end
  end
end

#get_producer ⇒ Object 



355
356
357
358
359
360
361
362
363
364
365
366
367
368
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 355

def get_producer
  if @share_producer
    @shared_producer
  else
    @producers_mutex.synchronize {
      producer = @producers[Thread.current.object_id]
      unless producer
        producer = @rdkafka.producer
        @producers[Thread.current.object_id] = producer
      end
      producer
    }
  end
end

#multi_workers_ready? ⇒ Boolean

Returns:

  • (Boolean) 


310
311
312
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 310

def multi_workers_ready?
  true
end

#setup_formatter(conf) ⇒ Object 



370
371
372
373
374
375
376
377
378
379
380
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 370

def setup_formatter(conf)
  type = conf['@type']
  case type
  when 'ltsv'
    require 'ltsv'
    Proc.new { |tag, time, record| LTSV.dump(record) }
  else
    @formatter = formatter_create(usage: 'rdkafka-plugin', conf: conf)
    @formatter.method(:format)
  end
end

#shutdown ⇒ Object 



324
325
326
327
328
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 324

def shutdown
  super
  wait_writing_threads
  shutdown_producers
end

#shutdown_producers ⇒ Object 



330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 330

def shutdown_producers
  if @share_producer
    close_producer(@shared_producer)
    @shared_producer = nil
  else
    @producers_mutex.synchronize {
      shutdown_threads = @producers.map { |key, producer|
        th = Thread.new {
          close_producer(producer)
        }
        th.abort_on_exception = true
        th
      }
      shutdown_threads.each { |th| th.join }
      @producers = {}
    }
  end
end

#start ⇒ Object 



299
300
301
302
303
304
305
306
307
308
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 299

def start
  if @share_producer
    @shared_producer = @rdkafka.producer
  else
    @producers = {}
    @producers_mutex = Mutex.new
  end

  super
end

#wait_writing_threads ⇒ Object 



314
315
316
317
318
319
320
321
322
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 314

def wait_writing_threads
  done = false
  until done do
    @writing_threads_mutex.synchronize do
      done = true if @writing_threads.empty?
    end
    sleep(1) unless done
  end
end

#write(chunk) ⇒ Object 



382
383
384
385
386
387
388
389
390
391
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
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
 
# File 'lib/fluent/plugin/out_rdkafka2.rb', line 382

def write(chunk)
  @writing_threads_mutex.synchronize { @writing_threads.add(Thread.current) }
  tag = chunk.metadata.tag
  topic = if @topic
            extract_placeholders(@topic, chunk)
          else
            (chunk.metadata.variables && chunk.metadata.variables[@topic_key_sym]) || @default_topic || tag
          end

  handlers = []

  headers = @headers.clone

  begin
    producer = get_producer
    chunk.msgpack_each { |time, record|
      begin
        record = inject_values_to_record(tag, time, record)
        record.delete(@topic_key) if @exclude_topic_key
        partition = (@exclude_partition ? record.delete(@partition_key) : record[@partition_key]) || @default_partition
        message_key = (@exclude_message_key ? record.delete(@message_key_key) : record[@message_key_key]) || @default_message_key

        @headers_from_record_accessors.each do |key, header_accessor|
          headers[key] = header_accessor.call(record)
        end

        unless @exclude_fields.empty?
          @exclude_field_accessors.each do |exclude_field_acessor|
            exclude_field_acessor.delete(record)
          end
        end

        record = @record_field_accessor.call(record) unless @record_field_accessor.nil?
        record_buf = @formatter_proc.call(tag, time, record)
        record_buf_bytes = record_buf.bytesize
        if @max_send_limit_bytes && record_buf_bytes > @max_send_limit_bytes
          log.warn "record size exceeds max_send_limit_bytes. Skip event:", :time => time, :record_size => record_buf_bytes
          log.debug "Skipped event:", :record => record
          next
        end
      rescue StandardError => e
        log.warn "unexpected error during format record. Skip broken event:", :error => e.to_s, :error_class => e.class.to_s, :time => time, :record => record
        next
      end

      handler = enqueue_with_retry(producer, topic, record_buf, message_key, partition, headers, time)
      if @rdkafka_delivery_handle_poll_timeout != 0
        handlers << handler
      end
    }
    handlers.each { |handler|
      handler.wait(max_wait_timeout: @rdkafka_delivery_handle_poll_timeout)
    }
  end
rescue Exception => e
  if @discard_kafka_delivery_failed
    log.warn "Delivery failed. Discard events:", :error => e.to_s, :error_class => e.class.to_s, :tag => tag
  else
	if @discard_kafka_delivery_failed_regex != nil && @discard_kafka_delivery_failed_regex.match?(e.to_s)
      log.warn "Delivery failed and matched regexp pattern #{@discard_kafka_delivery_failed_regex}. Discard events:", :error => e.to_s, :error_class => e.class.to_s, :tag => tag
	else
	  log.warn "Send exception occurred: #{e} at #{e.backtrace.first}"
      # Raise exception to retry sendind messages
      raise e
	end
  end
ensure
  @writing_threads_mutex.synchronize { @writing_threads.delete(Thread.current) }
end