Class: Fluent::DatadogOutput

Inherits:
Plugin::Output
  • Object
show all
Defined in:
lib/fluent/plugin/out_datadog.rb

Defined Under Namespace

Classes: DatadogClient, DatadogHTTPClient, DatadogTCPClient, RetryableError

Constant Summary collapse

DD_MAX_BATCH_LENGTH =

Max limits for transport regardless of Fluentd buffer, respecting https://docs.datadoghq.com/api/?lang=bash#logs

500
DD_MAX_BATCH_SIZE = 
5000000
DD_TRUNCATION_SUFFIX = 
"...TRUNCATED..."
DD_DEFAULT_SITE = 
"datadoghq.com"
DD_DEFAULT_HTTP_HOST_PREFIX = 
"http-intake.logs."
DD_DEFAULT_HTTP_ENDPOINT = 
"#{DD_DEFAULT_HTTP_HOST_PREFIX}#{DD_DEFAULT_SITE}".freeze
DD_DEFAULT_TCP_ENDPOINT = 
"intake.logs.datadoghq.com"
DEFAULT_BUFFER_TYPE = 
"memory"

Instance Method Summary collapse

Constructor Details

#initialize ⇒ DatadogOutput

Returns a new instance of DatadogOutput.



83
84
85
 
# File 'lib/fluent/plugin/out_datadog.rb', line 83

def initialize
  super
end

Instance Method Details

#batch_http_events(encoded_events, max_batch_length, max_request_size) ⇒ Object

Group HTTP events in batches



205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
 
# File 'lib/fluent/plugin/out_datadog.rb', line 205

def batch_http_events(encoded_events, max_batch_length, max_request_size)
  batches = []
  current_batch = []
  current_batch_size = 0
  encoded_events.each_with_index do |encoded_event, i|
    current_event_size = encoded_event.bytesize
    # If this unique log size is bigger than the request size, truncate it
    if current_event_size > max_request_size
      encoded_event = truncate(encoded_event, max_request_size)
      current_event_size = encoded_event.bytesize
    end

    if (i > 0 and i % max_batch_length == 0) or (current_batch_size + current_event_size > max_request_size)
      batches << current_batch
      current_batch = []
      current_batch_size = 0
    end

    current_batch_size += encoded_event.bytesize
    current_batch << encoded_event
  end
  batches << current_batch
  batches
end

#configure(conf) ⇒ Object 



87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
 
# File 'lib/fluent/plugin/out_datadog.rb', line 87

def configure(conf)
  compat_parameters_convert(conf, :buffer)
  super

  # Derive default host from `site` only for HTTP transport.
  # TCP users with a non-default site must set `host` explicitly;
  # TCP users with no host and the default site get the legacy TCP endpoint.
  if @host.nil? || @host.empty?
    if @use_http
      @host = "#{DD_DEFAULT_HTTP_HOST_PREFIX}#{@site}"
    elsif @site == DD_DEFAULT_SITE
      @host = DD_DEFAULT_TCP_ENDPOINT
    else
      # TCP + non-default site: we cannot safely derive a TCP intake hostname
      # from `site` (the HTTP-intake prefix is HTTP-only), and leaving @host
      # nil would surface as a cryptic connect-time error. Fail fast at
      # configure time with an actionable message.
      raise Fluent::ConfigError,
        "`host` is required when `use_http false` is combined with a non-default `site` " \
        "(`site #{@site.inspect}`). Set `host` explicitly to your TCP intake (e.g. " \
        "`intake.logs.#{@site}`)."
    end
  end

  return if @dd_hostname

  # Set dd_hostname if not already set (can be set when using fluentd as aggregator)
  @dd_hostname = %x[hostname -f 2> /dev/null].strip
  @dd_hostname = Socket.gethostname if @dd_hostname.empty?

  @timestamp_key = nil if nilish?(@timestamp_key)
end

#enrich_record(tag, time, record) ⇒ Object

Enrich records with metadata such as service, tags or source



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
 
# File 'lib/fluent/plugin/out_datadog.rb', line 250

def enrich_record(tag, time, record)
  if @dd_sourcecategory
    record["ddsourcecategory"] ||= @dd_sourcecategory
  end
  if @dd_source
    record["ddsource"] ||= @dd_source
  end
  if @dd_tags
    record["ddtags"] ||= @dd_tags
  end
  if @service
    record["service"] ||= @service
  end
  if @dd_hostname
    # set the record hostname to the configured dd_hostname only
    # if the record hostname is empty, ensuring having a hostname set
    # even if the record doesn't contain any.
    record["hostname"] ||= @dd_hostname
  end

  if @include_tag_key
    record[@tag_key] = tag
  end
  # If @timestamp_key already exists, we don't overwrite it.
  if @timestamp_key and record[@timestamp_key].nil? and time
    record[@timestamp_key] = Time.at(time).utc.iso8601(3)
  end

  container_tags = get_container_tags(record)
  unless container_tags.empty?
    if record["ddtags"].nil? || record["ddtags"].empty?
      record["ddtags"] = container_tags
    else
      record["ddtags"] = record["ddtags"] + "," + container_tags
    end
  end

  if @delete_extracted_tag_attributes
    record.delete('kubernetes')
    record.delete('docker')
  end

  record
end

#format(tag, time, record) ⇒ Object

This method is called when an event reaches Fluentd.



143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
 
# File 'lib/fluent/plugin/out_datadog.rb', line 143

def format(tag, time, record)
  # When Fluent::EventTime is msgpack'ed it gets converted to int with seconds
  # precision only. We explicitly convert it to floating point number, which
  # is compatible with Time.at below.
  record = enrich_record(tag, time.to_f, record)
  if @use_http
    record = JSON.generate(record)
  else
    if @use_json
      record = "#{api_key} #{JSON.generate(record)}"
    else
      record = "#{api_key} #{record}"
    end
  end
  [record].to_msgpack
end

#format_http_event_batch(events) ⇒ Object

Format batch of http events



245
246
247
 
# File 'lib/fluent/plugin/out_datadog.rb', line 245

def format_http_event_batch(events)
  "[#{events.join(',')}]"
end

#formatted_to_msgpack_binary? ⇒ Boolean

Returns:

  • (Boolean) 


124
125
126
 
# File 'lib/fluent/plugin/out_datadog.rb', line 124

def formatted_to_msgpack_binary?
  true
end

#get_container_tags(record) ⇒ Object

Collect docker and kubernetes tags for your logs using filter_kubernetes_metadata plugin, for more information about the attribute names, check: https://github.com/fabric8io/fluent-plugin-kubernetes_metadata_filter/blob/master/lib/fluent/plugin/filter_kubernetes_metadata.rb#L265



485
486
487
488
489
490
 
# File 'lib/fluent/plugin/out_datadog.rb', line 485

def get_container_tags(record)
  [
      get_kubernetes_tags(record),
      get_docker_tags(record)
  ].compact.join(",")
end

#get_docker_tags(record) ⇒ Object 



506
507
508
509
510
511
512
513
514
 
# File 'lib/fluent/plugin/out_datadog.rb', line 506

def get_docker_tags(record)
  if record.key?('docker') and not record.fetch('docker').nil?
    docker = record['docker']
    tags = Array.new
    tags.push("container_id:" + docker['container_id']) unless docker['container_id'].nil?
    return tags.join(",")
  end
  nil
end

#get_kubernetes_tags(record) ⇒ Object 



492
493
494
495
496
497
498
499
500
501
502
503
504
 
# File 'lib/fluent/plugin/out_datadog.rb', line 492

def get_kubernetes_tags(record)
  if record.key?('kubernetes') and not record.fetch('kubernetes').nil?
    kubernetes = record['kubernetes']
    tags = Array.new
    tags.push("image_name:" + kubernetes['container_image']) unless kubernetes['container_image'].nil?
    tags.push("container_name:" + kubernetes['container_name']) unless kubernetes['container_name'].nil?
    tags.push("kube_namespace:" + kubernetes['namespace_name']) unless kubernetes['namespace_name'].nil?
    tags.push("pod_name:" + kubernetes['pod_name']) unless kubernetes['pod_name'].nil?
    tags.push("container_id:" + kubernetes['docker_id']) unless kubernetes['docker_id'].nil?
    return tags.join(",")
  end
  nil
end

#gzip_compress(payload, compression_level) ⇒ Object

Compress logs with GZIP



296
297
298
299
300
301
302
303
304
305
306
 
# File 'lib/fluent/plugin/out_datadog.rb', line 296

def gzip_compress(payload, compression_level)
  gz = StringIO.new
  gz.set_encoding("BINARY")
  z = Zlib::GzipWriter.new(gz, compression_level)
  begin
    z.write(payload)
  ensure
    z.close
  end
  gz.string
end

#max(a, b) ⇒ Object 



240
241
242
 
# File 'lib/fluent/plugin/out_datadog.rb', line 240

def max(a, b)
  a > b ? a : b
end

#multi_workers_ready? ⇒ Boolean

Returns:

  • (Boolean) 


120
121
122
 
# File 'lib/fluent/plugin/out_datadog.rb', line 120

def multi_workers_ready?
  true
end

#new_client(logger, api_key, use_http, use_ssl, no_ssl_validation, host, ssl_port, port, http_proxy, custom_headers, use_compression, force_v1_routes) ⇒ Object

Build a new transport client



309
310
311
312
313
314
315
 
# File 'lib/fluent/plugin/out_datadog.rb', line 309

def new_client(logger, api_key, use_http, use_ssl, no_ssl_validation, host, ssl_port, port, http_proxy, custom_headers, use_compression, force_v1_routes)
  if use_http
    DatadogHTTPClient.new logger, use_ssl, no_ssl_validation, host, ssl_port, port, http_proxy, custom_headers, use_compression, api_key, force_v1_routes
  else
    DatadogTCPClient.new logger, use_ssl, no_ssl_validation, host, ssl_port, port
  end
end

#process_http_events(events, use_compression, compression_level, max_retries, max_backoff, max_batch_length, max_batch_size) ⇒ Object

Process and send a set of http events. Potentially break down this set of http events in smaller batches



185
186
187
188
189
190
191
192
193
194
 
# File 'lib/fluent/plugin/out_datadog.rb', line 185

def process_http_events(events, use_compression, compression_level, max_retries, max_backoff, max_batch_length, max_batch_size)
  batches = batch_http_events(events, max_batch_length, max_batch_size)
  batches.each do |batched_event|
    formatted_events = format_http_event_batch(batched_event)
    if use_compression
      formatted_events = gzip_compress(formatted_events, compression_level)
    end
    @client.send_retries(formatted_events, max_retries, max_backoff)
  end
end

#process_tcp_event(event, max_retries, max_backoff, max_batch_size) ⇒ Object

Process and send a single tcp event



197
198
199
200
201
202
 
# File 'lib/fluent/plugin/out_datadog.rb', line 197

def process_tcp_event(event, max_retries, max_backoff, max_batch_size)
  if event.bytesize > max_batch_size
    event = truncate(event, max_batch_size)
  end
  @client.send_retries(event, max_retries, max_backoff)
end

#shutdown ⇒ Object 



133
134
135
 
# File 'lib/fluent/plugin/out_datadog.rb', line 133

def shutdown
  super
end

#start ⇒ Object 



128
129
130
131
 
# File 'lib/fluent/plugin/out_datadog.rb', line 128

def start
  super
  @client = new_client(log, @api_key, @use_http, @use_ssl, @no_ssl_validation, @host, @ssl_port, @port, @http_proxy, @custom_headers, @use_compression, @force_v1_routes)
end

#terminate ⇒ Object 



137
138
139
140
 
# File 'lib/fluent/plugin/out_datadog.rb', line 137

def terminate
  super
  @client.close if @client
end

#truncate(event, max_length) ⇒ Object

Truncate events over the provided max length, appending a marker when truncated



231
232
233
234
235
236
237
238
 
# File 'lib/fluent/plugin/out_datadog.rb', line 231

def truncate(event, max_length)
  if event.length > max_length
    event = event[0..max_length - 1]
    event[max(0, max_length - DD_TRUNCATION_SUFFIX.length)..max_length - 1] = DD_TRUNCATION_SUFFIX
    return event
  end
  event
end

#write(chunk) ⇒ Object

NOTE! This method is called by internal thread, not Fluentd's main thread. 'chunk' is a buffer chunk that includes multiple formatted events.



163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
 
# File 'lib/fluent/plugin/out_datadog.rb', line 163

def write(chunk)
  begin
    if @use_http
      events = Array.new
      chunk.msgpack_each do |record|
        next if record.empty?
        events.push record[0]
      end
      process_http_events(events, @use_compression, @compression_level, @max_retries, @max_backoff, DD_MAX_BATCH_LENGTH, DD_MAX_BATCH_SIZE)
    else
      chunk.msgpack_each do |record|
        next if record.empty?
        process_tcp_event(record[0], @max_retries, @max_backoff, DD_MAX_BATCH_SIZE)
      end
    end
  rescue StandardError => e
    log.error("Processing exception in datadog forwarder #{e.class}: #{e.message}")
    raise
  end
end