Create :'elasticsearch.perform_health_check' config

To access the cluster name to assign to the
database_name attribute on the Elasticsearch datastore segment,
we currently make a request to the '_cluster/health' endpoint. This can
cause CPU and network spikes when large clusters start.
To avoid this performance issue, the
`:'elasticsearch.perform_health_check` config value can be set to false.
This will set the database_name attribute on the Elasticsearch datastore
segments to nil.

CHANGELOG.md

@@ -1,5 +1,17 @@
 # New Relic Ruby Agent Release Notes
 
+## v9.7.0
+
+Version 9.7.0 adds a configuration value that toggles whether the agent will perform health check requests on Elasticsearch clusters.
+
+- **Feature: Add elasticsearch.perform_health_check configuraiton**
+
+ In [#2360](https://github.com/newrelic/newrelic-ruby-agent/pull/2360), it was brought to our attention that large clusters starting the agent can be overwhelmed by requests to `_cluster/health` performed by the agent to acquire the cluster name. The cluster name, once acquired, is assigned to the `database_name` attribute on Elasticsearch datastore segments. Now, users can disable these requests using a new configuration value. When set to `false`, the `database_name` will be `nil`. Thank you [@erikkessler1](https://github.com/erikkessler1) for bringing this to our attention.
+
+ | Configuration name | Default | Behavior |
+ | --------------------------- | ------- | ------------------------------------------------------ |
+ | `elasticsearch.perform_health_check` | `true` | If true, the agent will query the cluster's `_cluster/health` endpoint to determine information about the cluster, including its name. When false, the database_name attribute will be `nil``. |
+
 ## v9.6.0
 
 Version 9.6.0 adds instrumentation for Async::HTTP, Ethon, and HTTPX, adds the ability to ignore specific routes with Roda, gleans Docker container IDs from cgroups v2-based containers, records additional synthetics attributes, fixes an issue with Rails 7.1 that could cause duplicate log records to be sent to New Relic, fixes a deprecation warning for the Sidekiq error handler, adds additional attributes for OpenTelemetry compatibility, and resolves some technical debt, thanks to the community.

lib/new_relic/agent/configuration/default_source.rb

@@ -313,6 +313,7 @@ def self.enforce_fallback(allowed_values: nil, fallback: nil)
  'webpacker:compile'
  ].join(',').freeze
 
+ # rubocop:disable Metrics/CollectionLiteralLength
  DEFAULTS = {
  # Critical
  :agent_enabled => {
@@ -1325,6 +1326,13 @@ def self.enforce_fallback(allowed_values: nil, fallback: nil)
  :allowed_from_server => true,
  :description => 'If `true`, the agent obfuscates Elasticsearch queries in transaction traces.'
  },
+ :'elasticsearch.perform_health_check' => {
+ :default => true,
+ :public => true,
+ :type => Boolean,
+ :allowed_from_server => true,
+ :description => "If true, the agent will query the cluster's `_cluster/health` endpoint to determine information about the cluster, including its name. When false, the database_name attribute for Elasticsearch datastore segements will be nil."
+ },
  # Heroku
  :'heroku.use_dyno_names' => {
  :default => true,
@@ -2405,6 +2413,7 @@ def self.enforce_fallback(allowed_values: nil, fallback: nil)
  :description => 'This value represents the total amount of memory available to the host (not the process), in mebibytes (1024 squared or 1,048,576 bytes).'
  }
  }.freeze
+ # rubocop:enable Metrics/CollectionLiteralLength
  end
  end
 end

lib/new_relic/agent/instrumentation/elasticsearch/instrumentation.rb

@@ -52,11 +52,13 @@ def nr_reported_query(query)
  end
 
  def nr_cluster_name
- return @nr_cluster_name if @nr_cluster_name
+ return @nr_cluster_name if defined?(:@nr_cluster_name)
  return if nr_hosts.empty?
 
  NewRelic::Agent.disable_all_tracing do
- @nr_cluster_name ||= perform_request('GET', '_cluster/health').body['cluster_name']
+ @nr_cluster_name ||= if NewRelic::Agent.config[:'elasticsearch.perform_health_check']
+ perform_request('GET', '_cluster/health').body['cluster_name']
+ end
  end
  rescue StandardError => e
  NewRelic::Agent.logger.error('Failed to get cluster name for elasticsearch', e)

test/multiverse/suites/elasticsearch/elasticsearch_instrumentation_test.rb

@@ -86,6 +86,26 @@ def test_segment_database_name
  assert_equal 'docker-cluster', @segment.database_name
  end
 
+ def test_segment_database_name_unknown_if_config_set
+ with_config(:'elasticsearch.perform_health_check' => false) do
+ # We need to create a new client for this test because
+ # the one defined in #setup is created while the
+ # :elasticsearch.perform_health_check config is true
+ # so @nr_cluster_name would already be set
+ @client = ::Elasticsearch::Client.new(
+ log: false,
+ hosts: "localhost:#{port}"
+ )
+
+ @client.index(index: 'my-index', id: 1, body: {title: 'Test'})
+ @client.indices.refresh(index: 'my-index')
+
+ search
+
+ assert_nil @segment.database_name
+ end
+ end
+
  def test_nosql_statement_recorded_params_obfuscated
  with_config(:'elasticsearch.obfuscate_queries' => true) do
  txn = in_transaction do