Class: Fluent::Plugin::KubernetesMetadataFilter

Inherits:
Filter
  • Object
show all
Includes:
KubernetesMetadata::CacheStrategy, KubernetesMetadata::Common, KubernetesMetadata::WatchNamespaces, KubernetesMetadata::WatchPods
Defined in:
lib/fluent/plugin/filter_kubernetes_metadata.rb

Overview

rubocop:disable Metrics/ClassLength

Constant Summary collapse

K8_POD_CA_CERT = 
'ca.crt'
K8_POD_TOKEN = 
'token'
REGEX_VAR_LOG_PODS =

rubocop:disable Layout/LineLength

'(var\.log\.pods)\.(?<namespace>[^_]+)_(?<pod_name>[a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*)_(?<pod_uuid>[a-z0-9-]*)\.(?<container_name>.+)\..*\.log$'
REGEX_VAR_LOG_CONTAINERS =

rubocop:disable Layout/LineLength

'(var\.log\.containers)\.(?<pod_name>[a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*)_(?<namespace>[^_]+)_(?<container_name>.+)-(?<docker_id>[a-z0-9]{64})\.log$'

Instance Method Summary collapse

Methods included from KubernetesMetadata::WatchPods

#get_pods_and_start_watcher, #process_pod_watcher_notices, #reset_pod_watch_retry_stats, #set_up_pod_thread, #start_pod_watch

Methods included from KubernetesMetadata::Common

#match_annotations, #parse_namespace_metadata, #parse_pod_metadata, #syms_to_strs

Methods included from KubernetesMetadata::WatchNamespaces

#get_namespaces_and_start_watcher, #process_namespace_watcher_notices, #reset_namespace_watch_retry_stats, #set_up_namespace_thread, #start_namespace_watch

Methods included from KubernetesMetadata::CacheStrategy

#get_pod_metadata

Constructor Details

#initialize ⇒ KubernetesMetadataFilter

Returns a new instance of KubernetesMetadataFilter.



173
174
175
176
177
178
 
# File 'lib/fluent/plugin/filter_kubernetes_metadata.rb', line 173

def initialize
  super
  @prev_time = Time.now
  @ssl_options = {}
  @auth_options = {}
end

Instance Method Details

#configure(conf) ⇒ Object

rubocop:disable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/MethodLength, Metrics/PerceivedComplexity



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
249
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
298
299
300
301
302
303
304
305
306
307
 
# File 'lib/fluent/plugin/filter_kubernetes_metadata.rb', line 180

def configure(conf) # rubocop:disable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/MethodLength, Metrics/PerceivedComplexity
  super

  require 'kubeclient'
  require 'lru_redux'

  @stats = KubernetesMetadata::Stats.new
  if @stats_interval <= 0
    @stats = KubernetesMetadata::NoOpStats.new
    define_singleton_method(:dump_stats) {} # rubocop:disable Lint/EmptyBlock
  end

  if @cache_ttl < 0
    log.info 'Setting the cache TTL to :none because it was <= 0'
    @cache_ttl = :none
  end

  # Caches pod/namespace UID tuples for a given container UID.
  @id_cache = LruRedux::TTL::ThreadSafeCache.new(@cache_size, @cache_ttl, @ignore_nil)

  # Use the container UID as the key to fetch a hash containing pod metadata
  @cache = LruRedux::TTL::ThreadSafeCache.new(@cache_size, @cache_ttl, @ignore_nil)

  # Use the namespace UID as the key to fetch a hash containing namespace metadata
  @namespace_cache = LruRedux::TTL::ThreadSafeCache.new(@cache_size, @cache_ttl, @ignore_nil)

  @tag_to_kubernetes_name_regexp_compiled = Regexp.compile(@tag_to_kubernetes_name_regexp)

  # Use Kubernetes default service account if we're in a pod.
  if @kubernetes_url.nil?
    log.debug('Kubernetes URL is not set - inspecting environ')
    env_host = ENV['KUBERNETES_SERVICE_HOST']
    env_port = ENV['KUBERNETES_SERVICE_PORT']

    if present?(env_host) && present?(env_port)
      if Resolv::IPv6::Regex.match?(env_host)
        # Brackets are needed around IPv6 addresses
        env_host = "[#{env_host}]"
      end
      @kubernetes_url = "https://#{env_host}:#{env_port}/api"
      log.debug("Kubernetes URL is now '#{@kubernetes_url}'")
    else
      log.debug('No Kubernetes URL could be found in config or environ')
    end
  end

  # Use SSL certificate and bearer token from Kubernetes service account.
  if Dir.exist?(@secret_dir)
    log.debug("Found directory with secrets: #{@secret_dir}")
    ca_cert = File.join(@secret_dir, K8_POD_CA_CERT)
    pod_token = File.join(@secret_dir, K8_POD_TOKEN)

    if !present?(@ca_file) && File.exist?(ca_cert)
      log.debug("Found CA certificate: #{ca_cert}")
      @ca_file = ca_cert
    end

    if !present?(@bearer_token_file) && File.exist?(pod_token)
      log.debug("Found pod token: #{pod_token}")
      @bearer_token_file = pod_token
    end
  end

  if present?(@kubernetes_url)
    @ssl_options = {
      client_cert: present?(@client_cert) ? OpenSSL::X509::Certificate.new(File.read(@client_cert)) : nil,
      client_key: present?(@client_key) ? OpenSSL::PKey::RSA.new(File.read(@client_key)) : nil,
      ca_file: @ca_file,
      verify_ssl: @verify_ssl ? OpenSSL::SSL::VERIFY_PEER : OpenSSL::SSL::VERIFY_NONE
    }

    if @ssl_partial_chain
      # taken from the ssl.rb OpenSSL::SSL::SSLContext code for DEFAULT_CERT_STORE
      require 'openssl'

      ssl_store = OpenSSL::X509::Store.new
      ssl_store.set_default_paths
      flagval = if defined? OpenSSL::X509::V_FLAG_PARTIAL_CHAIN
                  OpenSSL::X509::V_FLAG_PARTIAL_CHAIN
                else
                  # this version of ruby does not define OpenSSL::X509::V_FLAG_PARTIAL_CHAIN
                  0x80000
                end
      ssl_store.flags = OpenSSL::X509::V_FLAG_CRL_CHECK_ALL | flagval
      @ssl_options[:cert_store] = ssl_store
    end

    @auth_options[:bearer_token_file] = @bearer_token_file if present?(@bearer_token_file)

    create_client

    if @test_api_adapter
      log.info "Extending client with test api adapter #{@test_api_adapter}"
      @test_api_adapter = @test_api_adapter.gsub('::', '_')
                                           .gsub(/([A-Z]+)([A-Z][a-z])/, '\1_\2')
                                           .gsub(/([a-z\d])([A-Z])/, '\1_\2')
                                           .tr('-', '_')
                                           .downcase
      require_relative @test_api_adapter
      @client.extend(eval(@test_api_adapter)) # rubocop:disable Security/Eval
    end

    begin
      @client.api_valid?
    rescue KubeException => e
      raise Fluent::ConfigError, "Invalid Kubernetes API #{@apiVersion} endpoint #{@kubernetes_url}: #{e.message}"
    end

    if @watch
      if ENV['K8S_NODE_NAME'].nil? || ENV['K8S_NODE_NAME'].strip.empty?
        log.warn("!! The environment variable 'K8S_NODE_NAME' is not set to the node name which can affect the API server and watch efficiency !!") # rubocop:disable Layout/LineLength
      end

      pod_thread = Thread.new(self, &:set_up_pod_thread)
      pod_thread.abort_on_exception = true

      namespace_thread = Thread.new(self, &:set_up_namespace_thread)
      namespace_thread.abort_on_exception = true
    end
  end

  @annotations_regexps = []
  @annotation_match.each do |regexp|
    @annotations_regexps << Regexp.compile(regexp)
  rescue RegexpError => e
    log.error("Error: invalid regular expression in annotation_match: #{e}")
  end
end

#create_client ⇒ Object

rubocop:disable Metrics/MethodLength



309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
 
# File 'lib/fluent/plugin/filter_kubernetes_metadata.rb', line 309

def create_client # rubocop:disable Metrics/MethodLength
  log.debug('Creating K8S client')
  @client = nil
  @client = Kubeclient::Client.new(
    @kubernetes_url,
    @apiVersion,
    ssl_options: @ssl_options,
    auth_options: @auth_options,
    timeouts: {
      open: @open_timeout,
      read: @read_timeout
    },
    as: :parsed_symbolized
  )
end

#dump_stats ⇒ Object

rubocop:disable Metrics/AbcSize



131
132
133
134
135
136
137
138
139
140
141
142
143
144
 
# File 'lib/fluent/plugin/filter_kubernetes_metadata.rb', line 131

def dump_stats # rubocop:disable Metrics/AbcSize
  @curr_time = Time.now
  return if @curr_time.to_i - @prev_time.to_i < @stats_interval

  @prev_time = @curr_time
  @stats.set(:pod_cache_size, @cache.count)
  @stats.set(:namespace_cache_size, @namespace_cache.count) if @namespace_cache
  log.info(@stats)
  return unless log.level == Fluent::Log::LEVEL_TRACE

  log.trace("       id cache: #{@id_cache.to_a}")
  log.trace("      pod cache: #{@cache.to_a}")
  log.trace("namespace cache: #{@namespace_cache.to_a}")
end

#fetch_namespace_metadata(namespace_name) ⇒ Object

rubocop:disable Metrics/AbcSize, Metrics/MethodLength



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
 
# File 'lib/fluent/plugin/filter_kubernetes_metadata.rb', line 146

def fetch_namespace_metadata(namespace_name) # rubocop:disable Metrics/AbcSize, Metrics/MethodLength
  log.trace("fetching namespace metadata: #{namespace_name}")
  options = {
    resource_version: '0' # Fetch from API server cache instead of etcd quorum read
  }
  namespace_object = @client.get_namespace(namespace_name, nil, options)
  log.trace("raw metadata for #{namespace_name}: #{namespace_object}")
  metadata = parse_namespace_metadata(namespace_object)
  @stats.bump(:namespace_cache_api_updates)
  log.trace("parsed metadata for #{namespace_name}: #{metadata}")
  @namespace_cache[metadata['namespace_id']] = metadata
rescue KubeException => e
  if e.error_code == 401
    # recreate client to refresh token
    log.info("Encountered '401 Unauthorized' exception, recreating client to refresh token")
    create_client
  else
    log.error("Exception '#{e}' encountered fetching namespace metadata from Kubernetes API #{@apiVersion} endpoint #{@kubernetes_url}") # rubocop:disable Layout/LineLength
    @stats.bump(:namespace_cache_api_nil_error)
  end
  {}
rescue StandardError => e
  @stats.bump(:namespace_cache_api_nil_error)
  log.error("Exception '#{e}' encountered fetching namespace metadata from Kubernetes API #{@apiVersion} endpoint #{@kubernetes_url}") # rubocop:disable Layout/LineLength
  {}
end

#fetch_pod_metadata(namespace_name, pod_name) ⇒ Object

rubocop:disable Metrics/AbcSize, Metrics/MethodLength



101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
 
# File 'lib/fluent/plugin/filter_kubernetes_metadata.rb', line 101

def fetch_pod_metadata(namespace_name, pod_name) # rubocop:disable Metrics/AbcSize, Metrics/MethodLength
  log.trace("fetching pod metadata: #{namespace_name}/#{pod_name}")
  options = {
    resource_version: '0' # Fetch from API server cache instead of etcd quorum read
  }
  pod_object = @client.get_pod(pod_name, namespace_name, options)
  log.trace("raw metadata for #{namespace_name}/#{pod_name}: #{pod_object}")
  metadata = parse_pod_metadata(pod_object)
  @stats.bump(:pod_cache_api_updates)
  log.trace("parsed metadata for #{namespace_name}/#{pod_name}: #{metadata}")
  @cache[metadata['pod_id']] = metadata
rescue KubeException => e
  if e.error_code == 401
    # recreate client to refresh token
    log.info("Encountered '401 Unauthorized' exception, recreating client to refresh token")
    create_client
  elsif e.error_code == 404
    log.debug("Encountered '404 Not Found' exception, pod not found")
    @stats.bump(:pod_cache_api_nil_error)
  else
    log.error("Exception '#{e}' encountered fetching pod metadata from Kubernetes API #{@apiVersion} endpoint #{@kubernetes_url}") # rubocop:disable Layout/LineLength
    @stats.bump(:pod_cache_api_nil_error)
  end
  {}
rescue StandardError => e
  @stats.bump(:pod_cache_api_nil_error)
  log.error("Exception '#{e}' encountered fetching pod metadata from Kubernetes API #{@apiVersion} endpoint #{@kubernetes_url}") # rubocop:disable Layout/LineLength
  {}
end

#filter(tag, time, record) ⇒ Object

rubocop:disable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/MethodLength, Metrics/PerceivedComplexity



364
365
366
367
368
369
370
371
372
373
374
375
376
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/fluent/plugin/filter_kubernetes_metadata.rb', line 364

def filter(tag, time, record) # rubocop:disable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/MethodLength, Metrics/PerceivedComplexity
  tag_match_data = tag.match(@tag_to_kubernetes_name_regexp_compiled)
  batch_miss_cache = {}
  if tag_match_data
    cache_key = if tag_match_data.names.include?('pod_uuid') && !tag_match_data['pod_uuid'].nil?
                  tag_match_data['pod_uuid']
                else
                  tag_match_data['docker_id']
                end
    docker_id = tag_match_data.names.include?('docker_id') ? tag_match_data['docker_id'] : nil
    metadata = get_metadata_for_record(
      tag_match_data['namespace'],
      tag_match_data['pod_name'],
      tag_match_data['container_name'],
      cache_key,
      time,
      batch_miss_cache,
      docker_id
    )
  end
  if @lookup_from_k8s_field && record.key?('kubernetes') && record.key?('docker') &&
     record['kubernetes'].respond_to?(:has_key?) && record['docker'].respond_to?(:has_key?) &&
     record['kubernetes'].key?('namespace_name') &&
     record['kubernetes'].key?('pod_name') &&
     record['kubernetes'].key?('container_name') &&
     record['docker'].key?('container_id')
    k_metadata = get_metadata_for_record(
      record['kubernetes']['namespace_name'],
      record['kubernetes']['pod_name'],
      record['kubernetes']['container_name'],
      record['docker']['container_id'],
      time,
      batch_miss_cache,
      record['docker']['container_id']
    )
    metadata = k_metadata if k_metadata
  end
  dump_stats
  metadata ? record.merge(metadata) : record
end

#get_metadata_for_record(namespace_name, pod_name, container_name, cache_key, create_time, batch_miss_cache, docker_id) ⇒ Object

rubocop:disable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/MethodLength, Metrics/ParameterLists, Metrics/PerceivedComplexity



325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
 
# File 'lib/fluent/plugin/filter_kubernetes_metadata.rb', line 325

def get_metadata_for_record(namespace_name, pod_name, container_name, cache_key, create_time, batch_miss_cache, # rubocop:disable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/MethodLength, Metrics/ParameterLists, Metrics/PerceivedComplexity
                            docker_id)
  metadata = {
    'docker' => { 'container_id' => '' },
    'kubernetes' => {
      'container_name' => container_name,
      'namespace_name' => namespace_name,
      'pod_name' => pod_name
    }
  }
  metadata['docker']['container_id'] = docker_id unless docker_id.nil?
  container_cache_key = container_name
  if present?(@kubernetes_url)
    pod_metadata = get_pod_metadata(cache_key, namespace_name, pod_name, create_time, batch_miss_cache)
    if (pod_metadata.include? 'containers') && (pod_metadata['containers'].include? container_cache_key) && !@skip_container_metadata # rubocop:disable Layout/LineLength
      metadata['kubernetes']['container_image'] = pod_metadata['containers'][container_cache_key]['image']
      unless pod_metadata['containers'][container_cache_key]['image_id'].empty?
        metadata['kubernetes']['container_image_id'] =
          pod_metadata['containers'][container_cache_key]['image_id']
      end
      unless pod_metadata['containers'][container_cache_key]['containerID'].empty?
        metadata['docker']['container_id'] =
          pod_metadata['containers'][container_cache_key]['containerID']
      end
    end
    metadata['kubernetes'].merge!(pod_metadata) if pod_metadata
    metadata['kubernetes'].delete('containers')
  end
  metadata['kubernetes'].tap do |kube|
    kube.each_pair do |k, v|
      kube[k.dup] = v.dup
    end
  end
  if metadata['docker'] && (metadata['docker']['container_id'].nil? || metadata['docker']['container_id'].empty?)
    metadata.delete('docker')
  end
  metadata
end

#present?(object) ⇒ Boolean

copied from activesupport

Returns:

  • (Boolean) 


406
407
408
 
# File 'lib/fluent/plugin/filter_kubernetes_metadata.rb', line 406

def present?(object)
  object.respond_to?(:empty?) ? !object.empty? : !!object
end