Class: LogStash::Outputs::ElasticSearch::HttpClient::Pool

Inherits:
Object
  • Object
show all
Defined in:
lib/logstash/outputs/elasticsearch/http_client/pool.rb

Defined Under Namespace

Classes: BadResponseCodeError, HostUnreachableError, NoConnectionAvailableError

Constant Summary collapse

ROOT_URI_PATH = 
'/'.freeze
LICENSE_PATH = 
'/_license'.freeze
VERSION_7_TO_7_14 = 
::Gem::Requirement.new([">= 7.0.0", "< 7.14.0"])
DEFAULT_OPTIONS = 
{
  :healthcheck_path => ROOT_URI_PATH,
  :sniffing_path => "/_nodes/http",
  :bulk_path => "/_bulk",
  :scheme => 'http',
  :resurrect_delay => 5,
  :sniffing => false,
  :sniffer_delay => 10,
}.freeze
BUILD_FLAVOR_SERVERLESS = 
'serverless'.freeze
ELASTIC_API_VERSION = 
"Elastic-Api-Version".freeze
DEFAULT_EAV_HEADER = 
{ ELASTIC_API_VERSION => "2023-10-31" }.freeze
ES1_SNIFF_RE_URL = 
/\[([^\/]*)?\/?([^:]*):([0-9]+)\]/
ES2_AND_ABOVE_SNIFF_RE_URL = 
/([^\/]*)?\/?([^:]*):([0-9]+)/

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(logger, adapter, initial_urls = [], options = {}) ⇒ Pool

Returns a new instance of Pool.

Raises:

  • (ArgumentError) 


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
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 71

def initialize(logger, adapter, initial_urls=[], options={})
  @logger = logger
  @adapter = adapter
  @metric = options[:metric]
  @initial_urls = initial_urls

  raise ArgumentError, "No URL Normalizer specified!" unless options[:url_normalizer]
  @url_normalizer = options[:url_normalizer]
  DEFAULT_OPTIONS.merge(options).tap do |merged|
    @bulk_path = merged[:bulk_path]
    @sniffing_path = merged[:sniffing_path]
    @healthcheck_path = merged[:healthcheck_path]
    @resurrect_delay = merged[:resurrect_delay]
    @sniffing = merged[:sniffing]
    @sniffer_delay = merged[:sniffer_delay]
  end

  # Used for all concurrent operations in this class
  @state_mutex = Mutex.new

  # Holds metadata about all URLs
  @url_info = {}
  @stopping = false

  @license_checker = options[:license_checker] || LogStash::PluginMixins::ElasticSearch::NoopLicenseChecker::INSTANCE

  @last_es_version = Concurrent::AtomicReference.new
  @build_flavor = Concurrent::AtomicReference.new
end

Instance Attribute Details

#adapter ⇒ Object (readonly)

Returns the value of attribute adapter.



49
50
51
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 49

def adapter
  @adapter
end

#bulk_path ⇒ Object (readonly)

Returns the value of attribute bulk_path.



49
50
51
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 49

def bulk_path
  @bulk_path
end

#healthcheck_path ⇒ Object (readonly)

Returns the value of attribute healthcheck_path.



49
50
51
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 49

def healthcheck_path
  @healthcheck_path
end

#license_checker ⇒ Object (readonly)

license_checker is used by the pool specs



50
51
52
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 50

def license_checker
  @license_checker
end

#logger ⇒ Object (readonly)

Returns the value of attribute logger.



49
50
51
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 49

def logger
  @logger
end

#resurrect_delay ⇒ Object (readonly)

Returns the value of attribute resurrect_delay.



49
50
51
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 49

def resurrect_delay
  @resurrect_delay
end

#sniffer_delay ⇒ Object (readonly)

Returns the value of attribute sniffer_delay.



49
50
51
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 49

def sniffer_delay
  @sniffer_delay
end

#sniffing ⇒ Object (readonly)

Returns the value of attribute sniffing.



49
50
51
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 49

def sniffing
  @sniffing
end

#sniffing_path ⇒ Object (readonly)

Returns the value of attribute sniffing_path.



49
50
51
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 49

def sniffing_path
  @sniffing_path
end

Instance Method Details

#add_url(url) ⇒ Object 



407
408
409
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 407

def add_url(url)
  @url_info[url] ||= empty_url_meta
end

#address_str_to_uri(addr_str) ⇒ Object 



217
218
219
220
221
222
223
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 217

def address_str_to_uri(addr_str)
  matches = addr_str.match(ES1_SNIFF_RE_URL) || addr_str.match(ES2_AND_ABOVE_SNIFF_RE_URL)
  if matches
    host = matches[1].empty? ? matches[2] : matches[1]
    ::LogStash::Util::SafeURI.new("#{host}:#{matches[3]}")
  end
end

#alive_urls_count ⇒ Object 



138
139
140
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 138

def alive_urls_count
  @state_mutex.synchronize { @url_info.values.select {|v| v[:state] == :alive }.count }
end

#check_sniff ⇒ Object

Sniffs and returns the results. Does not update internal URLs!



192
193
194
195
196
197
198
199
200
201
202
203
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 192

def check_sniff
  _, url_meta, resp = perform_request(:get, @sniffing_path)
  @metric.increment(:sniff_requests)
  parsed = LogStash::Json.load(resp.body)
  nodes = parsed['nodes']
  if !nodes || nodes.empty?
    @logger.warn("Sniff returned no nodes! Will not update hosts.")
    return nil
  else
    sniff(nodes)
  end
end

#close ⇒ Object 



111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 111

def close
  @state_mutex.synchronize { @stopping = true }

  logger.debug  "Stopping sniffer"
  stop_sniffer

  logger.debug  "Stopping resurrectionist"
  stop_resurrectionist

  logger.debug  "Waiting for in use manticore connections"
  wait_for_in_use_connections

  logger.debug("Closing adapter #{@adapter}")
  @adapter.close
end

#empty_url_meta ⇒ Object 



415
416
417
418
419
420
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 415

def empty_url_meta
  {
    :in_use => 0,
    :state => :unknown
  }
end

#get_connection ⇒ Object 



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
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 461

def get_connection
  @state_mutex.synchronize do
    # The goal here is to pick a random connection from the least-in-use connections
    # We want some randomness so that we don't hit the same node over and over, but
    # we also want more 'fair' behavior in the event of high concurrency
    eligible_set = nil
    lowest_value_seen = nil
    @url_info.each do |url,meta|
      meta_in_use = meta[:in_use]
      next if meta[:state] == :dead

      if lowest_value_seen.nil? || meta_in_use < lowest_value_seen
        lowest_value_seen = meta_in_use
        eligible_set = [[url, meta]]
      elsif lowest_value_seen == meta_in_use
        eligible_set << [url, meta]
      end
    end

    return nil if eligible_set.nil?

    pick, pick_meta = eligible_set.sample
    pick_meta[:in_use] += 1

    [pick, pick_meta]
  end
end

#get_license(url) ⇒ Hash

Retrieve ES node license information

Parameters:

  • url (LogStash::Util::SafeURI) —

    ES node URL

Returns:

  • (Hash) —

    deserialized license document or empty Hash upon any error



244
245
246
247
248
249
250
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 244

def get_license(url)
  response = perform_request_to_url(url, :get, LICENSE_PATH)
  LogStash::Json.load(response.body)
rescue => e
  logger.error("Unable to get license information", url: url.sanitized.to_s, exception: e.class, message: e.message)
  {}
end

#get_root_path(url, params = {}) ⇒ Object 



315
316
317
318
319
320
321
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 315

def get_root_path(url, params={})
  resp = perform_request_to_url(url, :get, ROOT_URI_PATH, params)
  return resp, nil
rescue ::LogStash::Outputs::ElasticSearch::HttpClient::Pool::BadResponseCodeError => e
  logger.warn("Elasticsearch main endpoint returns #{e.response_code}", message: e.message, body: e.response_body)
  return nil, e
end

#health_check_request(url) ⇒ Object 



252
253
254
255
256
257
258
259
260
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 252

def health_check_request(url)
  logger.debug("Running health check to see if an Elasticsearch connection is working",
               :healthcheck_url => url.sanitized.to_s, :path => @healthcheck_path)
  response = perform_request_to_url(url, :head, @healthcheck_path)
  return response, nil
rescue ::LogStash::Outputs::ElasticSearch::HttpClient::Pool::BadResponseCodeError => e
  logger.warn("Health check failed", code: e.response_code, url: e.url, message: e.message)
  return nil, e
end

#healthcheck!(register_phase = true) ⇒ Object 



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
308
309
310
311
312
313
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 262

def healthcheck!(register_phase = true)
  # Try to keep locking granularity low such that we don't affect IO...
  @state_mutex.synchronize { @url_info.select {|url,meta| meta[:state] != :alive } }.each do |url,meta|
    begin
      _, health_bad_code_err = health_check_request(url)
      root_response, root_bad_code_err = get_root_path(url) if health_bad_code_err.nil? || register_phase

      # when called from resurrectionist skip the product check done during register phase
      if register_phase
        raise LogStash::ConfigurationError,
              "Could not read Elasticsearch. Please check the credentials" if root_bad_code_err&.invalid_credentials?
        raise LogStash::ConfigurationError,
              "Could not read Elasticsearch. Please check the privileges" if root_bad_code_err&.forbidden?
        # when customer_headers is invalid
        raise LogStash::ConfigurationError,
              "The Elastic-Api-Version header is not valid" if root_bad_code_err&.invalid_eav_header?
        # when it is not Elasticserach
        raise LogStash::ConfigurationError,
              "Could not connect to a compatible version of Elasticsearch" if root_bad_code_err.nil? && !elasticsearch?(root_response)

        test_serverless_connection(url, root_response)
      end

      raise health_bad_code_err if health_bad_code_err
      raise root_bad_code_err if root_bad_code_err

      # If no exception was raised it must have succeeded!
      if register_phase
        logger.info("Connected to ES instance", url: url.sanitized.to_s)
      else
        logger.warn("Restored connection to ES instance", url: url.sanitized.to_s)
      end

      # We check its ES version
      es_version, build_flavor = parse_es_version(root_response)
      logger.warn("Failed to retrieve Elasticsearch build flavor") if build_flavor.nil?
      logger.warn("Failed to retrieve Elasticsearch version data from connected endpoint, connection aborted", :url => url.sanitized.to_s) if es_version.nil?
      next if es_version.nil?

      @state_mutex.synchronize do
        meta[:version] = es_version
        set_last_es_version(es_version, url)
        set_build_flavor(build_flavor)

        alive = @license_checker.appropriate_license?(self, url)
        meta[:state] = alive ? :alive : :dead
      end
    rescue HostUnreachableError, BadResponseCodeError => e
      logger.warn("Attempted to resurrect connection to dead ES instance, but got an error", url: url.sanitized.to_s, exception: e.class, message: e.message)
    end
  end
end

#in_use_connections ⇒ Object 



134
135
136
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 134

def in_use_connections
  @state_mutex.synchronize { @url_info.values.select {|v| v[:in_use] > 0 } }
end

#last_es_version ⇒ Object 



496
497
498
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 496

def last_es_version
  @last_es_version.get
end

#major_version(version_string) ⇒ Object 



205
206
207
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 205

def major_version(version_string)
  version_string.split('.').first.to_i
end

#mark_dead(url, error) ⇒ Object 



441
442
443
444
445
446
447
448
449
450
451
452
453
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 441

def mark_dead(url, error)
  @state_mutex.synchronize do
    meta = @url_info[url]
    # In case a sniff happened removing the metadata just before there's nothing to mark
    # This is an extreme edge case, but it can happen!
    return unless meta
    logger.warn("Marking url as dead. Last error: [#{error.class}] #{error.message}",
                :url => url, :error_message => error.message, :error_class => error.class.name)
    meta[:state] = :dead
    meta[:last_error] = error
    meta[:last_errored_at] = Time.now
  end
end

#maximum_seen_major_version ⇒ Object 



500
501
502
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 500

def maximum_seen_major_version
  @state_mutex.synchronize { @maximum_seen_major_version }
end

#normalize_url(uri) ⇒ Object 



357
358
359
360
361
362
363
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 357

def normalize_url(uri)
  u = @url_normalizer.call(uri)
  if !u.is_a?(::LogStash::Util::SafeURI)
    raise "URL Normalizer returned a '#{u.class}' rather than a SafeURI! This shouldn't happen!"
  end
  u
end

#perform_request(method, path, params = {}, body = nil) ⇒ Object 



338
339
340
341
342
343
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 338

def perform_request(method, path, params={}, body=nil)
  with_connection do |url, url_meta|
    resp = perform_request_to_url(url, method, path, params, body)
    [url, url_meta, resp]
  end
end

#perform_request_to_url(url, method, path, params = {}, body = nil) ⇒ Object 



352
353
354
355
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 352

def perform_request_to_url(url, method, path, params={}, body=nil)
  params[:headers] = DEFAULT_EAV_HEADER.merge(params[:headers] || {}) if serverless?
  @adapter.perform_request(url, method, path, params, body)
end

#remove_url(url) ⇒ Object 



411
412
413
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 411

def remove_url(url)
  @url_info.delete(url)
end

#resurrectionist_alive? ⇒ Boolean

Returns:

  • (Boolean) 


334
335
336
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 334

def resurrectionist_alive?
  @resurrectionist ? @resurrectionist.alive? : nil
end

#return_connection(url) ⇒ Object 



489
490
491
492
493
494
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 489

def return_connection(url)
  @state_mutex.synchronize do
    info = @url_info[url]
    info[:in_use] -= 1 if info # Guard against the condition where the connection has already been deleted
  end
end

#serverless? ⇒ Boolean

Returns:

  • (Boolean) 


504
505
506
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 504

def serverless?
  @build_flavor.get == BUILD_FLAVOR_SERVERLESS
end

#size ⇒ Object 



403
404
405
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 403

def size
  @state_mutex.synchronize { @url_info.size }
end

#sniff(nodes) ⇒ Object 



209
210
211
212
213
214
215
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 209

def sniff(nodes)
  nodes.map do |id,info|
    # Skip master-only nodes
    next if info["roles"] && info["roles"] == ["master"]
    address_str_to_uri(info["http"]["publish_address"]) if info["http"]
  end.compact
end

#sniff! ⇒ Object

Sniffs the cluster then updates the internal URLs



185
186
187
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 185

def sniff!
  update_urls(check_sniff)
end

#sniffer_alive? ⇒ Boolean

Returns:

  • (Boolean) 


229
230
231
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 229

def sniffer_alive?
  @sniffer ? @sniffer.alive? : nil
end

#start ⇒ Object 



101
102
103
104
105
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 101

def start
  update_initial_urls
  start_resurrectionist
  start_sniffer if @sniffing
end

#start_resurrectionist ⇒ Object 



233
234
235
236
237
238
239
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 233

def start_resurrectionist
  @resurrectionist = Thread.new do
    until_stopped("resurrection", @resurrect_delay) do
      healthcheck!(false)
    end
  end
end

#start_sniffer ⇒ Object 



171
172
173
174
175
176
177
178
179
180
181
182
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 171

def start_sniffer
  @sniffer = Thread.new do
    until_stopped("sniffing", sniffer_delay) do
      begin
        sniff!
      rescue NoConnectionAvailableError => e
        @state_mutex.synchronize { # Synchronize around @url_info
          logger.warn("Elasticsearch output attempted to sniff for new connections but cannot. No living connections are detected. Pool contains the following current URLs", :url_info => @url_info) }
      end
    end
  end
end

#stop_resurrectionist ⇒ Object 



330
331
332
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 330

def stop_resurrectionist
  @resurrectionist.join if @resurrectionist
end

#stop_sniffer ⇒ Object 



225
226
227
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 225

def stop_sniffer
  @sniffer.join if @sniffer
end

#test_serverless_connection(url, root_response) ⇒ Object

Raises:

  • (LogStash::ConfigurationError) 


323
324
325
326
327
328
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 323

def test_serverless_connection(url, root_response)
  _, build_flavor = parse_es_version(root_response)
  params = { :headers => DEFAULT_EAV_HEADER }
  _, bad_code_err = get_root_path(url, params) if build_flavor == BUILD_FLAVOR_SERVERLESS
  raise LogStash::ConfigurationError, "The Elastic-Api-Version header is not valid" if bad_code_err&.invalid_eav_header?
end

#until_stopped(task_name, delay) ⇒ Object 



150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 150

def until_stopped(task_name, delay)
  last_done = Time.now
  until @state_mutex.synchronize { @stopping }
    begin
      now = Time.now
      if (now - last_done) >= delay
        last_done = now
        yield
      end
      sleep 1
    rescue => e
      logger.warn(
        "Error while performing #{task_name}",
        :error_message => e.message,
        :class => e.class.name,
        :backtrace => e.backtrace
      )
    end
  end
end

#update_initial_urls ⇒ Object 



107
108
109
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 107

def update_initial_urls
  update_urls(@initial_urls)
end

#update_urls(new_urls) ⇒ Object 



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
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 365

def update_urls(new_urls)
  return if new_urls.nil?

  # Normalize URLs
  new_urls = new_urls.map(&method(:normalize_url))

  # Used for logging nicely
  state_changes = {:removed => [], :added => []}
  @state_mutex.synchronize do
    # Add new connections
    new_urls.each do |url|
      # URI objects don't have real hash equality! So, since this isn't perf sensitive we do a linear scan
      unless @url_info.keys.include?(url)
        state_changes[:added] << url
        add_url(url)
      end
    end

    # Delete connections not in the new list
    @url_info.each do |url,_|
      unless new_urls.include?(url)
        state_changes[:removed] << url
        remove_url(url)
      end
    end
  end

  if state_changes[:removed].size > 0 || state_changes[:added].size > 0
    logger.info? && logger.info("Elasticsearch pool URLs updated", :changes => state_changes)
  end

  # Run an inline healthcheck anytime URLs are updated
  # This guarantees that during startup / post-startup
  # sniffing we don't have idle periods waiting for the
  # periodic sniffer to allow new hosts to come online
  healthcheck!
end

#url_info ⇒ Object 



142
143
144
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 142

def url_info
  @state_mutex.synchronize { @url_info }
end

#url_meta(url) ⇒ Object 



455
456
457
458
459
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 455

def url_meta(url)
  @state_mutex.synchronize do
    @url_info[url]
  end
end

#urls ⇒ Object 



146
147
148
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 146

def urls
  url_info.keys
end

#wait_for_in_use_connections ⇒ Object 



127
128
129
130
131
132
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 127

def wait_for_in_use_connections
  until in_use_connections.empty?
    logger.info "Blocked on shutdown to in use connections #{@state_mutex.synchronize {@url_info}}"
    sleep 1
  end
end

#with_connection ⇒ Object 



422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 422

def with_connection
  url, url_meta = get_connection

  # Custom error class used here so that users may retry attempts if they receive this error
  # should they choose to
  raise NoConnectionAvailableError, "No Available connections" unless url
  yield url, url_meta
rescue HostUnreachableError => e
  # Mark the connection as dead here since this is likely not transient
  mark_dead(url, e)
  raise e
rescue BadResponseCodeError => e
  # These aren't discarded from the pool because these are often very transient
  # errors
  raise e
ensure
  return_connection(url)
end