datadog 2.23.0 → 2.25.0

data/lib/datadog/core/remote/component.rb

@@ -11,9 +11,11 @@ module Datadog
   module Core
     module Remote
       # Configures the HTTP transport to communicate with the agent
-      # to fetch and sync the remote configuration
+      # to fetch and sync the remote configuration.
+      #
+      # @api private
       class Component
-        attr_reader :logger, :client, :healthy
+        attr_reader :logger, :client, :healthy, :worker
 
         def initialize(settings, capabilities, agent_settings, logger:)
           @logger = logger
@@ -23,7 +25,7 @@ module Datadog
 
           @barrier = Barrier.new(settings.remote.boot_timeout_seconds)
 
-          @client = Client.new(transport_v7, capabilities, logger: logger)
+          @client = Client.new(transport_v7, capabilities, settings: settings, logger: logger)
           @healthy = false
           logger.debug { "new remote configuration client: #{@client.id}" }
 
@@ -55,7 +57,7 @@ module Datadog
               end
 
               # client state is unknown, state might be corrupted
-              @client = Client.new(transport_v7, capabilities, logger: logger)
+              @client = Client.new(transport_v7, capabilities, settings: settings, logger: logger)
               @healthy = false
               logger.debug { "new remote configuration client: #{@client.id}" }
 

data/lib/datadog/core/remote/configuration/content.rb

@@ -22,6 +22,18 @@ module Datadog
           attr_accessor :version
 
           def initialize(path:, data:)
+            if data.nil?
+              # +data+ is passed to Digest calculation and also is
+              # unconditionally taken length of by +length+ method.
+              # As such, the class is not written to expect +data+ to be nil.
+              # Detect bad incoming values here to provide earlier diagnostics
+              # when developing tests, for example.
+              raise ArgumentError, 'data must not be nil'
+            end
+            unless String === data
+              raise ArgumentError, "Invalid type for data: #{data.class}: expected String"
+            end
+
             @path = path
             @data = data
             @apply_state = ApplyState::UNACKNOWLEDGED
@@ -72,8 +84,9 @@ module Datadog
           private_class_method :new
         end
 
-        # ContentList stores a list of Conetnt instances
-        # It provides convinient methods for finding content base on Configuration::Path and Configuration::Target
+        # ContentList stores a list of Content instances.
+        # It provides convenient methods for finding content based on
+        # Configuration::Path and Configuration::Target.
         class ContentList < Array
           class << self
             def parse(array)

data/lib/datadog/core/remote/configuration/digest.rb

@@ -24,10 +24,21 @@ module Datadog
           class InvalidHashTypeError < StandardError; end
           attr_reader :type, :hexdigest
 
-          DIGEST_CHUNK = 1024
-
           class << self
             def hexdigest(type, data)
+              unless String === data
+                # This class (Digest) passes +data+ to the Ruby standard
+                # library Digest routines without validating its type.
+                # The stdlib Digest requires a String, and the previous
+                # implementation of this class that used StringIO
+                # unconditionally read from +data+ without validating the
+                # type. Meaning, passing +nil+ as +data+ has never worked.
+                # It still doesn't work in the present implementation.
+                # Flag the nil data now to get earlier diagnostics when
+                # developing tests for example.
+                raise ArgumentError, "Invalid type for data: #{data.class}: expected String"
+              end
+
               d = case type
               when :sha256
                 ::Digest::SHA256.new
@@ -37,13 +48,9 @@ module Datadog
                 raise InvalidHashTypeError, type
               end
 
-              while (buf = data.read(DIGEST_CHUNK))
-                d.update(buf)
-              end
+              d.update(data)
 
               d.hexdigest
-            ensure
-              data.rewind
             end
           end
 

data/lib/datadog/core/remote/configuration/repository.rb

@@ -185,7 +185,7 @@ module Datadog
               end
             end
 
-            # Update existimng repository's contents
+            # Update existing repository's contents
             class Update
               attr_reader :path, :target, :content
 

data/lib/datadog/core/remote/configuration/target.rb

@@ -11,8 +11,15 @@ module Datadog
         class TargetMap < Hash
           class << self
             def parse(hash)
-              opaque_backend_state = hash['signed']['custom']['opaque_backend_state']
-              version = hash['signed']['version']
+              signed = hash.fetch('signed')
+              # Note that the +dig+ call permits +hash['signed']+ to be
+              # missing the +custom+ subtree entirely.
+              # Previously the subtree was required but +opaque_backend_state+
+              # could still be missing (and obtained here as nil).
+              opaque_backend_state = signed.dig('custom', 'opaque_backend_state')
+              # The version appears to be optional to the rest of this class,
+              # and we have tests that do not provide it.
+              version = signed['version']
 
               map = new
 
@@ -21,7 +28,7 @@ module Datadog
                 @version = version
               end
 
-              hash['signed']['targets'].each_with_object(map) do |(p, t), m|
+              signed.fetch('targets').each_with_object(map) do |(p, t), m|
                 path = Configuration::Path.parse(p)
                 target = Configuration::Target.parse(t)
 
@@ -46,9 +53,9 @@ module Datadog
         class Target
           class << self
             def parse(hash)
-              length = Integer(hash['length'])
-              digests = Configuration::DigestList.parse(hash['hashes'])
-              version = Integer(hash['custom']['v'])
+              length = Integer(hash.fetch('length'))
+              digests = Configuration::DigestList.parse(hash.fetch('hashes'))
+              version = Integer(hash.dig('custom', 'v'))
 
               new(digests: digests, length: length, version: version)
             end

data/lib/datadog/core/remote/transport/config.rb

@@ -2,6 +2,7 @@
 
 require_relative '../../../core/transport/request'
 require_relative '../../../core/transport/parcel'
+require_relative '../../../core/transport/transport'
 require_relative 'http/config'
 
 module Datadog
@@ -23,27 +24,13 @@ module Datadog
           end
 
           # Config transport
-          class Transport
-            attr_reader :client, :apis, :default_api, :current_api_id, :logger
-
-            def initialize(apis, default_api, logger: Datadog.logger)
-              @apis = apis
-              @logger = logger
-
-              @client = Remote::Transport::HTTP::Config::Client.new(current_api, logger: logger)
-            end
-
-            ##### there is only one transport! it's negotiation!
+          class Transport < Core::Transport::Transport
             def send_config(payload)
               json = JSON.dump(payload)
               parcel = EncodedParcel.new(json)
               request = Request.new(parcel)
 
-              @client.send_config_payload(request)
-            end
-
-            def current_api
-              @apis[HTTP::API::V7]
+              @client.send_request(:config, request)
             end
           end
         end

data/lib/datadog/core/remote/transport/http/config.rb

@@ -2,11 +2,10 @@
 
 require 'json'
 
-require_relative '../../../transport/http/client'
+require_relative '../../../transport/http/api/endpoint'
+require_relative '../../../transport/http/response'
 require_relative '../../../utils/base64'
 require_relative '../../../utils/truncation'
-require_relative '../../../transport/http/response'
-require_relative '../../../transport/http/api/endpoint'
 
 module Datadog
   module Core
@@ -17,7 +16,7 @@ module Datadog
           module Config
             # Response from HTTP transport for remote configuration
             class Response
-              include Datadog::Core::Transport::HTTP::Response
+              include Core::Transport::HTTP::Response
 
               def initialize(http_response, options = {}) # standard:disable Metrics/AbcSize,Metrics/CyclomaticComplexity,Metrics/MethodLength,Metrics/PerceivedComplexity
                 super(http_response)
@@ -101,7 +100,7 @@ module Datadog
 
                   {
                     path: h[:path].freeze,
-                    content: StringIO.new(content.freeze),
+                    content: content.freeze,
                   }
                 end.freeze
 
@@ -178,46 +177,7 @@ module Datadog
               end
             end
 
-            # Remote transport HTTP client
-            class Client < Core::Transport::HTTP::Client
-              def send_config_payload(request)
-                send_request(request) do |api, env|
-                  api.send_config(env)
-                end
-              end
-            end
-
             module API
-              # Extensions for HTTP API Spec
-              module Spec
-                attr_reader :config
-
-                def config=(endpoint)
-                  @config = endpoint
-                end
-
-                def send_config(env, &block)
-                  raise Core::Transport::HTTP::API::Spec::EndpointNotDefinedError.new('config', self) if config.nil?
-
-                  config.call(env, &block)
-                end
-              end
-
-              # Extensions for HTTP API Instance
-              module Instance
-                def send_config(env)
-                  unless spec.is_a?(Config::API::Spec)
-                    raise Core::Transport::HTTP::API::Instance::EndpointNotSupportedError.new(
-                      'config', self
-                    )
-                  end
-
-                  spec.send_config(env) do |request_env|
-                    call(request_env)
-                  end
-                end
-              end
-
               # Endpoint for remote configuration
               class Endpoint < Datadog::Core::Transport::HTTP::API::Endpoint
                 HEADER_CONTENT_TYPE = 'Content-Type'

data/lib/datadog/core/remote/transport/http/negotiation.rb

@@ -43,46 +43,7 @@ module Datadog
               attr_reader :version, :endpoints, :config, :span_events
             end
 
-            # Remote negotiation HTTP client
-            class Client < Core::Transport::HTTP::Client
-              def send_info_payload(request)
-                send_request(request) do |api, env|
-                  api.send_info(env)
-                end
-              end
-            end
-
             module API
-              # Extensions for HTTP API Spec
-              module Spec
-                attr_reader :info
-
-                def info=(endpoint)
-                  @info = endpoint
-                end
-
-                def send_info(env, &block)
-                  raise Core::Transport::HTTP::API::Spec::EndpointNotDefinedError.new('info', self) if info.nil?
-
-                  info.call(env, &block)
-                end
-              end
-
-              # Extensions for HTTP API Instance
-              module Instance
-                def send_info(env)
-                  unless spec.is_a?(Negotiation::API::Spec)
-                    raise Core::Transport::HTTP::API::Instance::EndpointNotSupportedError.new(
-                      'info', self
-                    )
-                  end
-
-                  spec.send_info(env) do |request_env|
-                    call(request_env)
-                  end
-                end
-              end
-
               # Endpoint for negotiation
               class Endpoint < Datadog::Core::Transport::HTTP::API::Endpoint
                 def initialize(path)

data/lib/datadog/core/remote/transport/http.rb

@@ -7,16 +7,6 @@ require_relative '../../transport/http'
 require_relative 'config'
 require_relative 'negotiation'
 
-# TODO: Improve negotiation to allow per endpoint selection
-#
-# Since endpoint negotiation happens at the `API::Spec` level there can not be
-# a mix of endpoints at various versions or versionless without describing all
-# the possible combinations as specs. See http/api.
-#
-# Below should be:
-# require_relative '../../transport/http/api'
-require_relative 'http/api'
-
 # TODO: Decouple transport/http
 #
 # Because a new transport is required for every (API, Client, Transport)
@@ -29,26 +19,30 @@ module Datadog
       module Transport
         # Namespace for HTTP transport components
         module HTTP
+          ROOT = Negotiation::API::Endpoint.new(
+            '/info',
+          )
+
+          V7 = Config::API::Endpoint.new(
+            '/v0.7/config',
+            Core::Encoding::JSONEncoder,
+          )
+
           module_function
 
           # Builds a new Transport::HTTP::Client with default settings
           # Pass a block to override any settings.
           def root(
             agent_settings:,
-            logger: Datadog.logger,
-            api_version: nil,
+            logger:,
             headers: nil
           )
             Core::Transport::HTTP.build(
-              api_instance_class: API::Instance,
               agent_settings: agent_settings,
               logger: logger,
-              api_version: api_version,
               headers: headers
             ) do |transport|
-              apis = API.defaults
-
-              transport.api API::ROOT, apis[API::ROOT]
+              transport.api 'root', ROOT
 
               # Call block to apply any customization, if provided
               yield(transport) if block_given?
@@ -59,20 +53,15 @@ module Datadog
           # Pass a block to override any settings.
           def v7(
             agent_settings:,
-            logger: Datadog.logger,
-            api_version: nil,
+            logger:,
             headers: nil
           )
             Core::Transport::HTTP.build(
-              api_instance_class: API::Instance,
               agent_settings: agent_settings,
               logger: logger,
-              api_version: api_version,
               headers: headers
             ) do |transport|
-              apis = API.defaults
-
-              transport.api API::V7, apis[API::V7]
+              transport.api 'v7', V7
 
               # Call block to apply any customization, if provided
               yield(transport) if block_given?

data/lib/datadog/core/remote/transport/negotiation.rb

@@ -1,6 +1,7 @@
 # frozen_string_literal: true
 
 require_relative '../../../core/transport/request'
+require_relative '../../../core/transport/transport'
 require_relative 'http/negotiation'
 
 # TODO: Resolve conceptual conundrum
@@ -32,24 +33,14 @@ module Datadog
           end
 
           # Negotiation transport
-          class Transport
-            attr_reader :client, :apis, :default_api, :current_api_id, :logger
-
-            def initialize(apis, default_api, logger: Datadog.logger)
-              @apis = apis
-              @logger = logger
-
-              @client = Remote::Transport::HTTP::Negotiation::Client.new(current_api, logger: logger)
-            end
-
-            def send_info
+          class Transport < Core::Transport::Transport
+            # TODO steep is complaining about this method, I tried to make
+            # it work but between module inclusions and steep diagnostics
+            # I don't understand what the problem is or what the steep wants.
+            def send_info # steep:ignore
               request = Request.new
 
-              @client.send_info_payload(request)
-            end
-
-            def current_api
-              @apis[HTTP::API::ROOT]
+              @client.send_request(:info, request)
             end
           end
         end

data/lib/datadog/core/semaphore.rb

@@ -20,10 +20,7 @@ module Datadog
 
       def wait(timeout = nil)
         wake_lock.synchronize do
-          # steep specifies that the second argument to wait is of type
-          # ::Time::_Timeout which for some reason is not Numeric and is not
-          # castable from Numeric.
-          wake.wait(wake_lock, timeout) # steep:ignore
+          wake.wait(wake_lock, timeout)
         end
       end
 

data/lib/datadog/core/telemetry/component.rb

@@ -14,14 +14,26 @@ require_relative '../utils/forking'
 module Datadog
   module Core
     module Telemetry
-      # Telemetry entrypoint, coordinates sending telemetry events at various points in app lifecycle.
-      # Note: Telemetry does not spawn its worker thread in fork processes, thus no telemetry is sent in forked processes.
+      # Telemetry entry point, coordinates sending telemetry events at
+      # various points in application lifecycle.
       #
       # @api private
       class Component
         ENDPOINT_COLLECTION_MESSAGE_LIMIT = 300
 
-        attr_reader :enabled, :logger, :transport, :worker
+        ONLY_ONCE = Utils::OnlyOnce.new
+
+        attr_reader :enabled
+        attr_reader :logger
+        attr_reader :transport
+        attr_reader :worker
+        attr_reader :settings
+        attr_reader :agent_settings
+        attr_reader :metrics_manager
+
+        # Alias for consistency with other components.
+        # TODO Remove +enabled+ method
+        alias_method :enabled?, :enabled
 
         include Core::Utils::Forking
         include Telemetry::Logging
@@ -50,6 +62,17 @@ module Datadog
           logger:,
           enabled:
         )
+          ONLY_ONCE.run do
+            Utils::AtForkMonkeyPatch.apply!
+
+            # All of the other at fork monkey patch callbacks reference
+            # globals, follow that pattern here to avoid having the component
+            # referenced via the at fork callbacks.
+            Datadog::Core::Utils::AtForkMonkeyPatch.at_fork(:child) do
+              Datadog.send(:components, allow_initialization: false)&.telemetry&.after_fork
+            end
+          end
+
           @enabled = enabled
           @log_collection_enabled = settings.telemetry.log_collection_enabled
           @logger = logger
@@ -110,7 +133,7 @@ module Datadog
         end
 
         def start(initial_event_is_change = false, components:)
-          return if !@enabled
+          return unless enabled?
 
           initial_event = if initial_event_is_change
             Event::SynthAppClientConfigurationChange.new(
@@ -136,44 +159,44 @@ module Datadog
         end
 
         def emit_closing!
-          return if !@enabled || forked?
+          return unless enabled?
 
           @worker.enqueue(Event::AppClosing.new)
         end
 
         def integrations_change!
-          return if !@enabled || forked?
+          return unless enabled?
 
           @worker.enqueue(Event::AppIntegrationsChange.new)
         end
 
         def log!(event)
-          return if !@enabled || forked? || !@log_collection_enabled
+          return unless enabled? && @log_collection_enabled
 
           @worker.enqueue(event)
         end
 
         # Wait for the worker to send out all events that have already
         # been queued, up to 15 seconds. Returns whether all events have
-        # been flushed.
+        # been flushed, or nil if telemetry is disabled.
         #
         # @api private
-        def flush
-          return if !@enabled || forked?
+        def flush(timeout: nil)
+          return unless enabled?
 
-          @worker.flush
+          @worker.flush(timeout: timeout)
         end
 
         # Report configuration changes caused by Remote Configuration.
         def client_configuration_change!(changes)
-          return if !@enabled || forked?
+          return unless enabled?
 
           @worker.enqueue(Event::AppClientConfigurationChange.new(changes, 'remote_config'))
         end
 
         # Report application endpoints
         def app_endpoints_loaded(endpoints, page_size: ENDPOINT_COLLECTION_MESSAGE_LIMIT)
-          return if !@enabled || forked?
+          return unless enabled?
 
           endpoints.each_slice(page_size).with_index do |endpoints_slice, i|
             @worker.enqueue(Event::AppEndpointsLoaded.new(endpoints_slice, is_first: i.zero?))
@@ -204,6 +227,22 @@ module Datadog
         def distribution(namespace, metric_name, value, tags: {}, common: true)
           @metrics_manager.distribution(namespace, metric_name, value, tags: tags, common: common)
         end
+
+        # When a fork happens, we generally need to do two things inside the
+        # child proess:
+        # 1. Restart the worker.
+        # 2. Discard any events and metrics that were submitted in the
+        #    parent process (because they will be sent out in the parent
+        #    process, sending them in the child would cause duplicate
+        #    submission).
+        def after_fork
+          # We cannot simply create a new instance of metrics manager because
+          # it is referenced from other objects (e.g. the worker).
+          # We must reset the existing instance.
+          @metrics_manager.clear
+
+          worker&.send(:after_fork_monkey_patched)
+        end
       end
     end
   end

data/lib/datadog/core/telemetry/event/app_started.rb

@@ -11,6 +11,12 @@ module Datadog
           def initialize(components:)
             # To not hold a reference to the component tree, generate
             # the event payload here in the constructor.
+            #
+            # Important: do not store data that contains (or is derived from)
+            # the runtime_id or sequence numbers.
+            # This event is reused when a process forks, but in the
+            # child process the runtime_id would be different and sequence
+            # number is reset.
             @configuration = configuration(components.settings, components.agent_settings)
             @install_signature = install_signature(components.settings)
             @products = products(components)
@@ -30,10 +36,19 @@ module Datadog
             }
           end
 
+          # Whether the event is actually the app-started event.
+          # For the app-started event we follow up by sending
+          # app-dependencies-loaded, if the event is
+          # app-client-configuration-change we don't send
+          # app-dependencies-loaded.
+          def app_started?
+            true
+          end
+
           private
 
           def products(components)
-            # @type var products: Hash[Symbol, Hash[Symbol, Hash[Symbol, String | Integer] | bool | nil]]
+            # @type var products: telemetry_products
             products = {
               appsec: {
                 # TODO take appsec status out of component tree?
@@ -159,6 +174,25 @@ module Datadog
               get_telemetry_origin(settings, 'tracing.contrib.peer_service_mapping')
             )
 
+            # OpenTelemetry configuration options (using environment variable names)
+            otel_exporter_headers_string = settings.opentelemetry.exporter.headers&.map { |key, value| "#{key}=#{value}" }&.join(',')
+            otel_exporter_metrics_headers_string = settings.opentelemetry.metrics.headers&.map { |key, value| "#{key}=#{value}" }&.join(',')
+            list.push(
+              conf_value('OTEL_EXPORTER_OTLP_ENDPOINT', settings.opentelemetry.exporter.endpoint, seq_id, get_telemetry_origin(settings, 'opentelemetry.exporter.endpoint')),
+              conf_value('OTEL_EXPORTER_OTLP_HEADERS', otel_exporter_headers_string, seq_id, get_telemetry_origin(settings, 'opentelemetry.exporter.headers')),
+              conf_value('OTEL_EXPORTER_OTLP_PROTOCOL', settings.opentelemetry.exporter.protocol, seq_id, get_telemetry_origin(settings, 'opentelemetry.exporter.protocol')),
+              conf_value('OTEL_EXPORTER_OTLP_TIMEOUT', settings.opentelemetry.exporter.timeout_millis, seq_id, get_telemetry_origin(settings, 'opentelemetry.exporter.timeout_millis')),
+              conf_value('DD_METRICS_OTEL_ENABLED', settings.opentelemetry.metrics.enabled, seq_id, get_telemetry_origin(settings, 'opentelemetry.metrics.enabled')),
+              conf_value('OTEL_METRICS_EXPORTER', settings.opentelemetry.metrics.exporter, seq_id, get_telemetry_origin(settings, 'opentelemetry.metrics.exporter')),
+              conf_value('OTEL_EXPORTER_OTLP_METRICS_ENDPOINT', settings.opentelemetry.metrics.endpoint, seq_id, get_telemetry_origin(settings, 'opentelemetry.metrics.endpoint')),
+              conf_value('OTEL_EXPORTER_OTLP_METRICS_HEADERS', otel_exporter_metrics_headers_string, seq_id, get_telemetry_origin(settings, 'opentelemetry.metrics.headers')),
+              conf_value('OTEL_EXPORTER_OTLP_METRICS_PROTOCOL', settings.opentelemetry.metrics.protocol, seq_id, get_telemetry_origin(settings, 'opentelemetry.metrics.protocol')),
+              conf_value('OTEL_EXPORTER_OTLP_METRICS_TIMEOUT', settings.opentelemetry.metrics.timeout_millis, seq_id, get_telemetry_origin(settings, 'opentelemetry.metrics.timeout_millis')),
+              conf_value('OTEL_EXPORTER_OTLP_METRICS_TEMPORALITY_PREFERENCE', settings.opentelemetry.metrics.temporality_preference, seq_id, get_telemetry_origin(settings, 'opentelemetry.metrics.temporality_preference')),
+              conf_value('OTEL_METRIC_EXPORT_INTERVAL', settings.opentelemetry.metrics.export_interval_millis, seq_id, get_telemetry_origin(settings, 'opentelemetry.metrics.export_interval_millis')),
+              conf_value('OTEL_METRIC_EXPORT_TIMEOUT', settings.opentelemetry.metrics.export_timeout_millis, seq_id, get_telemetry_origin(settings, 'opentelemetry.metrics.export_timeout_millis')),
+            )
+
             # Whitelist of configuration options to send in additional payload object
             TARGET_OPTIONS.each do |option_path|
               split_option = option_path.split('.')
@@ -243,6 +277,7 @@ module Datadog
           # - `default`: set when the user has not set any configuration for the key (defaults to a value)
           # - `unknown`: set for cases where it is difficult/not possible to determine the source of a config.
           def conf_value(name, value, seq_id, origin)
+            # @type var result: telemetry_configuration
             result = {name: name, value: value, origin: origin, seq_id: seq_id}
             if origin == 'fleet_stable_config'
               fleet_id = Core::Configuration::StableConfig.configuration.dig(:fleet, :id)