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_6_TO_7 = 
::Gem::Requirement.new([">= 6.0.0", "< 7.0.0"])
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_FLAVOUR_SERVERLESS = 
'serverless'.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) 


53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 53

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_flavour = Concurrent::AtomicReference.new
end

Instance Attribute Details

#adapter ⇒ Object (readonly)

Returns the value of attribute adapter.



32
33
34
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 32

def adapter
  @adapter
end

#bulk_path ⇒ Object (readonly)

Returns the value of attribute bulk_path.



32
33
34
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 32

def bulk_path
  @bulk_path
end

#healthcheck_path ⇒ Object (readonly)

Returns the value of attribute healthcheck_path.



32
33
34
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 32

def healthcheck_path
  @healthcheck_path
end

#license_checker ⇒ Object (readonly)

license_checker is used by the pool specs



33
34
35
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 33

def license_checker
  @license_checker
end

#logger ⇒ Object (readonly)

Returns the value of attribute logger.



32
33
34
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 32

def logger
  @logger
end

#resurrect_delay ⇒ Object (readonly)

Returns the value of attribute resurrect_delay.



32
33
34
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 32

def resurrect_delay
  @resurrect_delay
end

#sniffer_delay ⇒ Object (readonly)

Returns the value of attribute sniffer_delay.



32
33
34
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 32

def sniffer_delay
  @sniffer_delay
end

#sniffing ⇒ Object (readonly)

Returns the value of attribute sniffing.



32
33
34
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 32

def sniffing
  @sniffing
end

#sniffing_path ⇒ Object (readonly)

Returns the value of attribute sniffing_path.



32
33
34
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 32

def sniffing_path
  @sniffing_path
end

Instance Method Details

#add_url(url) ⇒ Object 



390
391
392
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 390

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

#address_str_to_uri(addr_str) ⇒ Object 



199
200
201
202
203
204
205
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 199

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 



120
121
122
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 120

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!



174
175
176
177
178
179
180
181
182
183
184
185
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 174

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 



93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 93

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

#elasticsearch?(url) ⇒ Boolean 



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/logstash/outputs/elasticsearch/http_client/pool.rb', line 278

def elasticsearch?(url)
  begin
    response = perform_request_to_url(url, :get, ROOT_URI_PATH)
  rescue ::LogStash::Outputs::ElasticSearch::HttpClient::Pool::BadResponseCodeError => e
    return false if response.code == 401 || response.code == 403
    raise e
  end

  version_info = LogStash::Json.load(response.body)
  return false if version_info['version'].nil?

  version = ::Gem::Version.new(version_info["version"]['number'])
  return false if version < ::Gem::Version.new('6.0.0')

  if VERSION_6_TO_7.satisfied_by?(version)
    return valid_tagline?(version_info)
  elsif VERSION_7_TO_7_14.satisfied_by?(version)
    build_flavor = version_info["version"]['build_flavor']
    return false if build_flavor.nil? || build_flavor != 'default' || !valid_tagline?(version_info)
  else
    # case >= 7.14
    lower_headers = response.headers.transform_keys {|key| key.to_s.downcase }
    product_header = lower_headers['x-elastic-product']
    return false if product_header != 'Elasticsearch'
  end
  return true
rescue => e
  logger.error("Unable to retrieve Elasticsearch version", url: url.sanitized.to_s, exception: e.class, message: e.message)
  false
end

#empty_url_meta ⇒ Object 



398
399
400
401
402
403
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 398

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

#get_connection ⇒ Object 



444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 444

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_es_version(url) ⇒ Object 



479
480
481
482
483
484
485
486
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 479

def get_es_version(url)
  response = perform_request_to_url(url, :get, ROOT_URI_PATH)
  return nil unless (200..299).cover?(response.code)

  response = LogStash::Json.load(response.body)

  response.fetch('version', {})
end

#get_license(url) ⇒ Hash

Retrieve ES node license information



226
227
228
229
230
231
232
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 226

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

#health_check_request(url) ⇒ Object

Raises:



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

def health_check_request(url)
  response = perform_request_to_url(url, :head, @healthcheck_path)
  raise BadResponseCodeError.new(response.code, url, nil, response.body) unless (200..299).cover?(response.code)
end

#healthcheck!(register_phase = true) ⇒ Object 



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
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 239

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
      logger.debug("Running health check to see if an Elasticsearch connection is working",
                    :healthcheck_url => url.sanitized.to_s, :path => @healthcheck_path)
      health_check_request(url)

      # when called from resurrectionist skip the product check done during register phase
      if register_phase
        if !elasticsearch?(url)
          raise LogStash::ConfigurationError, "Could not connect to a compatible version of Elasticsearch"
        end
      end
      # If no exception was raised it must have succeeded!
      logger.warn("Restored connection to ES instance", url: url.sanitized.to_s)
      # We reconnected to this node, check its ES version
      version_info = get_es_version(url)
      es_version = version_info.fetch('number', nil)
      build_flavour = version_info.fetch('build_flavor', nil)

      if es_version.nil?
        logger.warn("Failed to retrieve Elasticsearch version data from connected endpoint, connection aborted", :url => url.sanitized.to_s)
        next
      end
      @state_mutex.synchronize do
        meta[:version] = es_version
        set_last_es_version(es_version, url)
        set_build_flavour(build_flavour)

        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 



116
117
118
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 116

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

#last_es_version ⇒ Object 



488
489
490
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 488

def last_es_version
  @last_es_version.get
end

#major_version(version_string) ⇒ Object 



187
188
189
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 187

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

#mark_dead(url, error) ⇒ Object 



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

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 



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

def maximum_seen_major_version
  @state_mutex.synchronize { @maximum_seen_major_version }
end

#normalize_url(uri) ⇒ Object 



340
341
342
343
344
345
346
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 340

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 



322
323
324
325
326
327
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 322

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 



336
337
338
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 336

def perform_request_to_url(url, method, path, params={}, body=nil)
  @adapter.perform_request(url, method, path, params, body)
end

#remove_url(url) ⇒ Object 



394
395
396
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 394

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

#resurrectionist_alive? ⇒ Boolean 



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

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

#return_connection(url) ⇒ Object 



472
473
474
475
476
477
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 472

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 



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

def serverless?
  @build_flavour.get == BUILD_FLAVOUR_SERVERLESS
end

#size ⇒ Object 



386
387
388
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 386

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

#sniff(nodes) ⇒ Object 



191
192
193
194
195
196
197
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 191

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



167
168
169
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 167

def sniff!
  update_urls(check_sniff)
end

#sniffer_alive? ⇒ Boolean 



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

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

#start ⇒ Object 



83
84
85
86
87
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 83

def start
  update_initial_urls
  start_resurrectionist
  start_sniffer if @sniffing
end

#start_resurrectionist ⇒ Object 



215
216
217
218
219
220
221
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 215

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

#start_sniffer ⇒ Object 



153
154
155
156
157
158
159
160
161
162
163
164
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 153

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 



314
315
316
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 314

def stop_resurrectionist
  @resurrectionist.join if @resurrectionist
end

#stop_sniffer ⇒ Object 



207
208
209
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 207

def stop_sniffer
  @sniffer.join if @sniffer
end

#until_stopped(task_name, delay) ⇒ Object 



132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 132

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 



89
90
91
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 89

def update_initial_urls
  update_urls(@initial_urls)
end

#update_urls(new_urls) ⇒ Object 



348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 348

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 



124
125
126
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 124

def url_info
  @state_mutex.synchronize { @url_info }
end

#url_meta(url) ⇒ Object 



438
439
440
441
442
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 438

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

#urls ⇒ Object 



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

def urls
  url_info.keys
end

#valid_tagline?(version_info) ⇒ Boolean 



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

def valid_tagline?(version_info)
  tagline = version_info['tagline']
  tagline == "You Know, for Search"
end

#wait_for_in_use_connections ⇒ Object 



109
110
111
112
113
114
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 109

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 



405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
 
# File 'lib/logstash/outputs/elasticsearch/http_client/pool.rb', line 405

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