ddtrace 1.0.0 → 1.9.0

data/lib/datadog/tracing/flush.rb

@@ -3,71 +3,93 @@
 module Datadog
   module Tracing
     module Flush
-      # Consumes only completed traces (where all spans have finished)
-      class Finished
-        # Consumes and returns completed traces (where all spans have finished)
-        # from the provided \trace_op, if any.
+      # Consumes and returns a {TraceSegment} to be flushed, from
+      # the provided {TraceSegment}.
+      #
+      # Only finished spans are consumed. Any spans consumed are
+      # removed from +trace_op+ as a side effect. Unfinished spans are
+      # unaffected.
+      #
+      # @abstract
+      class Base
+        # Consumes and returns a {TraceSegment} to be flushed, from
+        # the provided {TraceSegment}.
         #
-        # Any traces consumed are removed from +trace_op+ as a side effect.
+        # Only finished spans are consumed. Any spans consumed are
+        # removed from +trace_op+ as a side effect. Unfinished spans are
+        # unaffected.
         #
+        # @param [TraceOperation] trace_op
         # @return [TraceSegment] trace to be flushed, or +nil+ if the trace is not finished
         def consume!(trace_op)
-          return unless full_flush?(trace_op)
+          return unless flush?(trace_op)
 
           get_trace(trace_op)
         end
 
-        def full_flush?(trace_op)
-          trace_op && trace_op.sampled? && trace_op.finished?
+        # Should we consume spans from the +trace_op+?
+        # @abstract
+        def flush?(trace_op)
+          raise NotImplementedError
         end
 
         protected
 
+        # Consumes all finished spans from trace.
+        # @return [TraceSegment]
         def get_trace(trace_op)
-          trace_op.flush!
+          trace_op.flush! do |spans|
+            spans.select! { |span| single_sampled?(span) } unless trace_op.sampled?
+
+            spans
+          end
+        end
+
+        # Single Span Sampling has chosen to keep this span
+        # regardless of the trace-level sampling decision
+        def single_sampled?(span)
+          span.get_metric(Sampling::Span::Ext::TAG_MECHANISM) == Sampling::Ext::Mechanism::SPAN_SAMPLING_RATE
+        end
+      end
+
+      # Consumes and returns completed traces (where all spans have finished),
+      # if any, from the provided +trace_op+.
+      #
+      # Spans consumed are removed from +trace_op+ as a side effect.
+      class Finished < Base
+        # Are all spans finished?
+        def flush?(trace_op)
+          trace_op && trace_op.finished?
         end
       end
 
-      # Performs partial trace flushing to avoid large traces residing in memory for too long
-      class Partial
+      # Consumes and returns completed or partially completed
+      # traces from the provided +trace_op+, if any.
+      #
+      # Partial trace flushing avoids large traces residing in memory for too long.
+      #
+      # Partially completed traces, where not all spans have finished,
+      # will only be returned if there are at least
+      # +@min_spans_for_partial+ finished spans.
+      #
+      # Spans consumed are removed from +trace_op+ as a side effect.
+      class Partial < Base
         # Start flushing partial trace after this many active spans in one trace
         DEFAULT_MIN_SPANS_FOR_PARTIAL_FLUSH = 500
 
         attr_reader :min_spans_for_partial
 
         def initialize(options = {})
+          super()
           @min_spans_for_partial = options.fetch(:min_spans_before_partial_flush, DEFAULT_MIN_SPANS_FOR_PARTIAL_FLUSH)
         end
 
-        # Consumes and returns completed or partially completed
-        # traces from the provided +trace_op+, if any.
-        #
-        # Partially completed traces, where not all spans have finished,
-        # will only be returned if there are at least
-        # +@min_spans_for_partial+ finished spans.
-        #
-        # Any spans consumed are removed from +trace_op+ as a side effect.
-        #
-        # @return [TraceSegment] partial or complete trace to be flushed, or +nil+ if no spans are finished
-        def consume!(trace_op)
-          return unless partial_flush?(trace_op)
-
-          get_trace(trace_op)
-        end
-
-        def partial_flush?(trace_op)
-          return false unless trace_op.sampled?
+        def flush?(trace_op)
           return true if trace_op.finished?
           return false if trace_op.finished_span_count < @min_spans_for_partial
 
           true
         end
-
-        protected
-
-        def get_trace(trace_op)
-          trace_op.flush!
-        end
       end
     end
   end

data/lib/datadog/tracing/metadata/analytics.rb

@@ -1,7 +1,7 @@
 # typed: true
 
-require 'datadog/tracing/analytics'
-require 'datadog/tracing/metadata/ext'
+require_relative '../analytics'
+require_relative 'ext'
 
 module Datadog
   module Tracing

data/lib/datadog/tracing/metadata/errors.rb

@@ -1,8 +1,8 @@
 # typed: false
 
-require 'datadog/core/error'
+require_relative '../../core/error'
 
-require 'datadog/tracing/metadata/ext'
+require_relative 'ext'
 
 module Datadog
   module Tracing

data/lib/datadog/tracing/metadata/ext.rb

@@ -20,6 +20,11 @@ module Datadog
         # Name of external service that performed the work
         TAG_PEER_SERVICE = 'peer.service'
 
+        TAG_KIND = 'span.kind'
+
+        # Set this tag to `1.0` if the span is a Service Entry span.
+        TAG_TOP_LEVEL = '_dd.top_level'
+
         # Defines constants for trace analytics
         # @public_api
         module Analytics
@@ -40,14 +45,22 @@ module Datadog
         # @public_api
         # Tags related to distributed tracing
         module Distributed
+          # What mechanism was used to make this trace's sampling decision.
+          # @see Datadog::Tracing::Sampling::Ext::Mechanism
+          TAG_DECISION_MAKER = '_dd.p.dm'
+
           TAG_ORIGIN = '_dd.origin'
           TAG_SAMPLING_PRIORITY = '_sampling_priority_v1'
+
+          # Trace tags with this prefix will propagate from a trace through distributed tracing.
+          # Distributed headers tags with this prefix will be injected into the active trace.
+          TAGS_PREFIX = '_dd.p.'
         end
 
         # @public_api
         module Errors
           STATUS = 1
-          TAG_MSG = 'error.msg'
+          TAG_MSG = 'error.message'
           TAG_STACK = 'error.stack'
           TAG_TYPE = 'error.type'
         end
@@ -58,11 +71,14 @@ module Datadog
           TAG_BASE_URL = 'http.base_url'
           TAG_METHOD = 'http.method'
           TAG_STATUS_CODE = 'http.status_code'
+          TAG_USER_AGENT = 'http.useragent'
           TAG_URL = 'http.url'
           TYPE_INBOUND = AppTypes::TYPE_WEB.freeze
           TYPE_OUTBOUND = 'http'
           TYPE_PROXY = 'proxy'
           TYPE_TEMPLATE = 'template'
+          TAG_CLIENT_IP = 'http.client_ip'
+          HEADER_USER_AGENT = 'User-Agent'
 
           # General header functionality
           module Headers
@@ -123,6 +139,8 @@ module Datadog
           TAG_HOSTNAME = '_dd.hostname'
           TAG_TARGET_HOST = 'out.host'
           TAG_TARGET_PORT = 'out.port'
+          TAG_DESTINATION_NAME = 'network.destination.name'
+          TAG_DESTINATION_PORT = 'network.destination.port'
         end
 
         # @public_api
@@ -145,6 +163,15 @@ module Datadog
           TYPE = 'sql'
           TAG_QUERY = 'sql.query'
         end
+
+        # @public_api
+        module SpanKind
+          TAG_SERVER = 'server'
+          TAG_CLIENT = 'client'
+          TAG_PRODUCER = 'producer'
+          TAG_CONSUMER = 'consumer'
+          TAG_INTERNAL = 'internal'
+        end
       end
     end
   end

data/lib/datadog/tracing/metadata/tagging.rb

@@ -1,8 +1,8 @@
 # typed: false
 
-require 'datadog/core/environment/ext'
+require_relative '../../core/environment/ext'
 
-require 'datadog/tracing/metadata/ext'
+require_relative 'ext'
 
 module Datadog
   module Tracing
@@ -65,11 +65,26 @@ module Datadog
           tags.each { |k, v| set_tag(k, v) }
         end
 
+        # Returns true if the provided `tag` was set to a non-nil value.
+        # False otherwise.
+        #
+        # @param [String] tag the tag or metric to check for presence
+        # @return [Boolean] if the tag is present and not nil
+        def has_tag?(tag) # rubocop:disable Naming/PredicateName
+          !get_tag(tag).nil? # nil is considered not present, thus we can't use `Hash#has_key?`
+        end
+
         # This method removes a tag for the given key.
         def clear_tag(key)
           meta.delete(key)
         end
 
+        # Convenient interface for setting a single tag.
+        alias []= set_tag
+
+        # Convenient interface for getting a single tag.
+        alias [] get_tag
+
         # Return the metric with the given key, nil if it doesn't exist.
         def get_metric(key)
           metrics[key] || meta[key]
@@ -95,6 +110,12 @@ module Datadog
           metrics.delete(key)
         end
 
+        # Returns a copy of all metadata.
+        # Keys for `@meta` and `@metrics` don't collide, by construction.
+        def tags
+          @meta.merge(@metrics)
+        end
+
         protected
 
         def meta

data/lib/datadog/tracing/metadata.rb

@@ -1,8 +1,8 @@
 # typed: true
 
-require 'datadog/tracing/metadata/analytics'
-require 'datadog/tracing/metadata/tagging'
-require 'datadog/tracing/metadata/errors'
+require_relative 'metadata/analytics'
+require_relative 'metadata/tagging'
+require_relative 'metadata/errors'
 
 module Datadog
   module Tracing

data/lib/datadog/tracing/pipeline/span_filter.rb

@@ -1,7 +1,7 @@
 # typed: true
 
 require 'set'
-require 'datadog/tracing/pipeline/span_processor'
+require_relative 'span_processor'
 
 module Datadog
   module Tracing
@@ -15,17 +15,21 @@ module Datadog
       #
       # @public_api
       class SpanFilter < SpanProcessor
-        # NOTE: this SpanFilter implementation only handles traces in which child spans appear
-        # after parent spans in the trace array. If in the future child spans can be before
+        # NOTE: This SpanFilter implementation only handles traces in which child spans appear
+        # before parent spans in the trace array. If in the future child spans can be after
         # parent spans, then the code below will need to be updated.
         # @!visibility private
         def call(trace)
           deleted = Set.new
 
-          trace.spans.delete_if do |span|
+          span_count = trace.spans.length
+          trace.spans.reverse_each.with_index do |span, i|
             should_delete = deleted.include?(span.parent_id) || drop_it?(span)
-            deleted << span.id if should_delete
-            should_delete
+
+            if should_delete
+              deleted << span.id
+              trace.spans.delete_at(span_count - 1 - i)
+            end
           end
 
           trace

data/lib/datadog/tracing/pipeline.rb

@@ -1,9 +1,9 @@
 # typed: true
 
-require 'datadog/core'
+require_relative '../core'
 
-require 'datadog/tracing/pipeline/span_filter'
-require 'datadog/tracing/pipeline/span_processor'
+require_relative 'pipeline/span_filter'
+require_relative 'pipeline/span_processor'
 
 module Datadog
   module Tracing

data/lib/datadog/tracing/propagation/http.rb

@@ -1,109 +1,14 @@
 # typed: false
 
-require 'datadog/core'
-
-require 'datadog/tracing/configuration/ext'
-require 'datadog/tracing/sampling/ext'
-require 'datadog/tracing/distributed/headers/b3'
-require 'datadog/tracing/distributed/headers/b3_single'
-require 'datadog/tracing/distributed/headers/datadog'
-
-require 'datadog/tracing/trace_digest'
-require 'datadog/tracing/trace_operation'
+require_relative '../contrib/http/distributed/propagation'
 
 module Datadog
   module Tracing
     module Propagation
       # Propagation::HTTP helps extracting and injecting HTTP headers.
+      # DEV-2.0: This file has been moved to Contrib. Should be deleted in the next release.
       # @public_api
-      module HTTP
-        PROPAGATION_STYLES = {
-          Configuration::Ext::Distributed::PROPAGATION_STYLE_B3 => Distributed::Headers::B3,
-          Configuration::Ext::Distributed::PROPAGATION_STYLE_B3_SINGLE_HEADER => Distributed::Headers::B3Single,
-          Configuration::Ext::Distributed::PROPAGATION_STYLE_DATADOG => Distributed::Headers::Datadog
-        }.freeze
-
-        # inject! popolates the env with span ID, trace ID and sampling priority
-        def self.inject!(digest, env)
-          # Prevent propagation from being attempted if trace headers provided are nil.
-          if digest.nil?
-            Datadog.logger.debug(
-              'Cannot inject trace headers into env to propagate over HTTP: trace headers are nil.'.freeze
-            )
-            return
-          end
-
-          digest = digest.to_digest if digest.is_a?(TraceOperation)
-
-          # Inject all configured propagation styles
-          Datadog.configuration.tracing.distributed_tracing.propagation_inject_style.each do |style|
-            propagator = PROPAGATION_STYLES[style]
-            begin
-              propagator.inject!(digest, env) unless propagator.nil?
-            rescue => e
-              Datadog.logger.error(
-                'Error injecting propagated trace headers into the environment. ' \
-                "Cause: #{e} Location: #{Array(e.backtrace).first}"
-              )
-            end
-          end
-        end
-
-        # extract returns trace headers containing the span ID, trace ID and
-        # sampling priority defined in env.
-        def self.extract(env)
-          trace_digest = nil
-          dd_trace_digest = nil
-
-          Datadog.configuration.tracing.distributed_tracing.propagation_extract_style.each do |style|
-            propagator = PROPAGATION_STYLES[style]
-            next if propagator.nil?
-
-            # Extract trace headers
-            # DEV: `propagator.extract` will return `nil`, where `Propagation::HTTP#extract` will not
-            begin
-              extracted_trace_digest = propagator.extract(env)
-            rescue => e
-              Datadog.logger.error(
-                'Error extracting propagated trace headers from the environment. ' \
-                "Cause: #{e} Location: #{Array(e.backtrace).first}"
-              )
-            end
-
-            # Skip this style if no valid headers were found
-            next if extracted_trace_digest.nil?
-
-            # Keep track of the Datadog extract trace headers, we want to return
-            #   this one if we have one
-            if extracted_trace_digest && style == Configuration::Ext::Distributed::PROPAGATION_STYLE_DATADOG
-              dd_trace_digest = extracted_trace_digest
-            end
-
-            # No previously extracted trace headers, use the one we just extracted
-            if trace_digest.nil?
-              trace_digest = extracted_trace_digest
-            else
-              unless trace_digest.trace_id == extracted_trace_digest.trace_id \
-                      && trace_digest.span_id == extracted_trace_digest.span_id
-                # Return an empty/new trace headers if we have a mismatch in values extracted
-                msg = "#{trace_digest.trace_id} != #{extracted_trace_digest.trace_id} && " \
-                      "#{trace_digest.span_id} != #{extracted_trace_digest.span_id}"
-                Datadog.logger.debug(
-                  "Cannot extract trace headers from HTTP: extracted trace headers differ, #{msg}"
-                )
-                # DEV: This will return from `self.extract` not this `each` block
-                return TraceDigest.new
-              end
-            end
-          end
-
-          # Return the extracted trace headers if we found one or else a new empty trace headers
-          # Always return the Datadog trace headers if one exists since it has more
-          #   information than the B3 headers e.g. origin, expanded priority
-          #   sampling values, etc
-          dd_trace_digest || trace_digest || nil
-        end
-      end
+      HTTP = Tracing::Contrib::HTTP::Distributed::Propagation.new
     end
   end
 end

data/lib/datadog/tracing/runtime/metrics.rb

@@ -1,6 +1,6 @@
 # typed: true
 
-require 'datadog/core'
+require_relative '../../core'
 
 module Datadog
   module Tracing

data/lib/datadog/tracing/sampling/all_sampler.rb

@@ -1,6 +1,6 @@
 # typed: true
 
-require 'datadog/tracing/sampling/sampler'
+require_relative 'sampler'
 
 module Datadog
   module Tracing

data/lib/datadog/tracing/sampling/ext.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 # typed: strict
 
 module Datadog
@@ -21,6 +23,35 @@ module Datadog
           # through the {Datadog::Tracing::Sampling::RuleSampler}.
           USER_KEEP = 2
         end
+
+        # List of what mechanism was used to make the trace-level sampling decision.
+        module Mechanism
+          # Single Span Sampled.
+          SPAN_SAMPLING_RATE = 8
+        end
+
+        # List of how the decision was made for the trace-level sampling.
+        #
+        # These values used to populate the {Datadog::Tracing::Metadata::Ext::Distributed::TAG_DECISION_MAKER} tag.
+        #
+        # The decision has two parts, separated by a `-`:
+        # `part1-sampling_mechanism`. `part1` is currently not populated, thus
+        # this tag is currently formatted as `"-sampling_mechanism"`.
+        module Decision
+          # Used before the tracer receives any rates from agent and there are no rules configured.
+          DEFAULT = '-0'
+          # The sampling rate received in the agent's http response.
+          AGENT_RATE = '-1'
+          # Sampling rule or sampling rate based on tracer config.
+          TRACE_SAMPLING_RULE = '-3'
+          # User directly sets sampling priority via {Tracing.reject!} or {Tracing.keep!},
+          # or by a custom sampler implementation.
+          MANUAL = '-4'
+          # Formerly AppSec.
+          ASM = '-5'
+          # Single Span Sampled.
+          SPAN_SAMPLING_RATE = '-8'
+        end
       end
     end
   end

data/lib/datadog/tracing/sampling/priority_sampler.rb

@@ -1,20 +1,16 @@
 # typed: true
 
-require 'forwardable'
-
-require 'datadog/tracing/sampling/ext'
-require 'datadog/tracing/sampling/all_sampler'
-require 'datadog/tracing/sampling/rate_sampler'
-require 'datadog/tracing/sampling/rate_by_service_sampler'
+require_relative 'ext'
+require_relative 'all_sampler'
+require_relative 'rate_sampler'
+require_relative 'rate_by_service_sampler'
 
 module Datadog
   module Tracing
     module Sampling
-      # {Datadog:::Tracing::Sampling::PrioritySampler}
+      # {Datadog::Tracing::Sampling::PrioritySampler}
       # @public_api
       class PrioritySampler
-        extend Forwardable
-
         # NOTE: We do not advise using a pre-sampler. It can save resources,
         # but pre-sampling at rates < 100% may result in partial traces, unless
         # the pre-sampler knows exactly how to drop a span without dropping its ancestors.
@@ -25,17 +21,26 @@ module Datadog
 
         def initialize(opts = {})
           @pre_sampler = opts[:base_sampler] || AllSampler.new
-          @priority_sampler = opts[:post_sampler] || RateByServiceSampler.new
+          @priority_sampler = opts[:post_sampler] || RateByServiceSampler.new(decision: Sampling::Ext::Decision::AGENT_RATE)
         end
 
         def sample?(trace)
           @pre_sampler.sample?(trace)
         end
 
+        # DEV-2.0:We should get rid of this complicated interaction between @pre_sampler and @priority_sampler.
+        # DEV-2.0:If the user wants to configure a custom sampler, we should only allow them to provide a complete
+        # DEV-2.0:sampling suite, not having this convoluted support for mixing arbitrary provided samplers in
+        # DEV-2.0:the PrioritySampler. Ideally, the PrioritySampler is only used by Datadog.
+        # DEV-2.0:There are too many edge cases and combinations to work around currently in this class.
         def sample!(trace)
+          # The priority that was set before the sampler ran.
+          # This comes from distributed tracing priority propagation.
+          distributed_sampling_priority = priority_assigned?(trace)
+
           # If pre-sampling is configured, do it first. (By default, this will sample at 100%.)
           # NOTE: Pre-sampling at rates < 100% may result in partial traces; not recommended.
-          trace.sampled = pre_sample?(trace) ? @pre_sampler.sample!(trace) : true
+          trace.sampled = pre_sample?(trace) ? preserving_priority_sampling(trace) { @pre_sampler.sample!(trace) } : true
 
           if trace.sampled?
             # If priority sampling has already been applied upstream, use that value.
@@ -57,9 +62,39 @@ module Datadog
           end
 
           trace.sampled?
+        ensure
+          if trace.sampling_priority && trace.sampling_priority > 0
+            # Don't modify decision if priority was set upstream.
+            if !distributed_sampling_priority && !trace.has_tag?(Tracing::Metadata::Ext::Distributed::TAG_DECISION_MAKER)
+              # If no sampling priority being assigned at this point, a custom
+              # sampler implementation is configured: this means the user has
+              # full control over the sampling decision.
+              trace.set_tag(
+                Tracing::Metadata::Ext::Distributed::TAG_DECISION_MAKER,
+                Sampling::Ext::Decision::MANUAL
+              )
+            end
+          else
+            # The sampler decided to not keep this span, removing sampling decision.
+            trace.clear_tag(Tracing::Metadata::Ext::Distributed::TAG_DECISION_MAKER)
+          end
         end
 
-        def_delegators :@priority_sampler, :update
+        # (see Datadog::Tracing::Sampling::RateByServiceSampler#update)
+        def update(rate_by_service, decision: nil)
+          @priority_sampler.update(rate_by_service, decision: decision)
+        end
+
+        # Check if the Priority Sampling decision is to keep or drop the trace.
+        # Other factors can influence the sampling decision; this method is only
+        # responsible for interpreting the Sampling Priority decision.
+        #
+        # @param priority_sampling [Integer] priority sampling number
+        # @return [Boolean] true if trace is "kept" by priority sampling
+        # @return [Boolean] false if trace is "dropped" by priority sampling
+        def self.sampled?(priority_sampling)
+          priority_sampling >= Ext::Priority::AUTO_KEEP
+        end
 
         private
 
@@ -84,6 +119,23 @@ module Datadog
           end
         end
 
+        # Ensures the trace's priority sampling decision is not changed by the @pre_sampler.
+        # The @pre_sampler should only change `trace.sampled`.
+        def preserving_priority_sampling(trace)
+          sampling_priority = trace.sampling_priority
+          sampling_decision = trace.get_tag(Tracing::Metadata::Ext::Distributed::TAG_DECISION_MAKER)
+
+          yield.tap do
+            trace.sampling_priority = sampling_priority
+
+            if sampling_decision
+              trace.set_tag(Tracing::Metadata::Ext::Distributed::TAG_DECISION_MAKER, sampling_decision)
+            else
+              trace.clear_tag(Tracing::Metadata::Ext::Distributed::TAG_DECISION_MAKER)
+            end
+          end
+        end
+
         # Ensures the trace is always propagated to the writer and that
         # the sample rate metric represents the true client-side sampling.
         def preserving_sampling(trace)