jaeger-client 0.10.0 → 1.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 7de16fd42664a20ad343e3fdc701f057b93df9e9c8ac6e76d4ea6b36895a7733
-  data.tar.gz: 32421b715a39b2710cae5bf9f7c6ca075872717ac53964de98429f063ebc93ba
+  metadata.gz: 16e6a368a65ce75432681a4487909971193d06e6bca4aff72cf39b0a4af2eb8f
+  data.tar.gz: f6a665ce70c4804484fefb1fb5d1c0ddbd41b117b78d688559fa9adceaa55e1c
 SHA512:
-  metadata.gz: dacf1771b57e8f12074a0426b75b09b6a19610c13d4cb86b687e9930fc24a7e4a20f5c5e2819760e02a88baf01a8438f2175688be57dff2dffd80e5d45303522
-  data.tar.gz: b8bdaa192c6c2109b7711431ec9eebe3e4352e5e71426c65355abd5aad358cf0f4ceaaecb25dec2cf3a622d0982294ea578858234c71682a841f7bb3bb4090c0
+  metadata.gz: 06165b0c04dd22c36064cd4f4e829c06e8883dc56b5b9b48aa8e4c82c673a150dd0d426bb56ffc73148d466965c11d1518c322dedf4134e986e54c7bfaad24c7
+  data.tar.gz: 2e91ace1219ad1f9456098ef897c234c92f79b2190794c95775a56a8b79086869bed55e4f4076d69d7aea7b0630fa78be5d6811ed4bd04a865ffba2d8105f3c8

data/.github/workflows/ci.yml

@@ -0,0 +1,33 @@
+name: Ruby
+
+on:
+  push:
+  pull_request:
+    branches: [ master ]
+  schedule:
+    # Runs at 00:00 UTC on the 1st of every month.
+    - cron: '0 0 1 * *'
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        ruby-version: ['2.7.2', '3.0', '3.1']
+
+    services:
+      postgres:
+        image: docker
+    steps:
+    - uses: actions/checkout@v2
+      with:
+        submodules: true
+    - name: Set up Ruby
+      uses: ruby/setup-ruby@v1
+      with:
+        ruby-version: ${{ matrix.ruby-version }}
+        bundler-cache: true
+    - name: Run tests
+      run: bundle exec rake
+    - name: Run crossdock
+      run: make crossdock

data/.rubocop.yml

@@ -1,8 +1,13 @@
+inherit_from: .rubocop_todo.yml
+
 require: rubocop-rspec
 
 AllCops:
   Exclude:
     - 'thrift/**/*'
+    - 'vendor/bundle/**/*'
+  NewCops: enable
+  TargetRubyVersion: '2.7'
 
 Style/Documentation:
   Enabled: no
@@ -19,6 +24,9 @@ RSpec/ExampleLength:
 RSpec/MultipleExpectations:
   Enabled: no
 
+RSpec/MultipleMemoizedHelpers:
+  Enabled: no
+
 RSpec/MessageSpies:
   Enabled: no
 
@@ -46,5 +54,8 @@ Style/FrozenStringLiteralComment:
   Include:
     - 'lib/**/*'
 
-Metrics/LineLength:
+Layout/LineLength:
   Max: 120
+
+Style/SingleLineMethods:
+  Enabled: false

data/.rubocop_todo.yml

@@ -0,0 +1,13 @@
+# This configuration was generated by
+# `rubocop --auto-gen-config`
+# on 2022-01-25 17:49:13 UTC using RuboCop version 1.25.0.
+# The point is for the user to remove these configuration records
+# one by one as the offenses are removed from the code base.
+# Note that changes in the inspected code, or installation of new
+# versions of RuboCop, may require this file to be generated again.
+
+# Offense count: 1
+# Cop supports --auto-correct.
+Style/CollectionCompact:
+  Exclude:
+    - 'lib/jaeger/tracer.rb'

data/README.md

@@ -1,7 +1,9 @@
 Jaeger::Client
 ================
 [![Gem Version](https://badge.fury.io/rb/jaeger-client.svg)](https://rubygems.org/gems/jaeger-client)
-[![Build Status](https://travis-ci.org/salemove/jaeger-client-ruby.svg)](https://travis-ci.org/salemove/jaeger-client-ruby)
+[![Build Status](https://github.com/salemove/jaeger-client-ruby/actions/workflows/ci.yml/badge.svg?branch=master)](https://github.com/salemove/jaeger-client-ruby/actions/workflows/ci.yml?query=branch%3Amaster)
+
+**This project is not actively maintained. Please consider using [OpenTelemetry](https://opentelemetry.io/)**
 
 OpenTracing Tracer implementation for Jaeger in Ruby
 
@@ -26,6 +28,9 @@ OpenTracing.start_active_span('span name') do
     # do something else
   end
 end
+
+# don't kill the program too soon, allow time for the background reporter to send the traces
+sleep 2
 ```
 
 See [opentracing-ruby](https://github.com/opentracing/opentracing-ruby) for more examples.
@@ -109,8 +114,8 @@ Set `sampler` to
   Jaeger::Samplers::PerOperation.new(
     strategies: {
       per_operation_strategies: [
-        { operation: 'GET /articles', probabilistic_sampling: 0.5 },
-        { operation: 'POST /articles', probabilistic_sampling: 1.0 }
+        { operation: 'GET /articles', probabilistic_sampling: { sampling_rate: 0.5 } },
+        { operation: 'POST /articles', probabilistic_sampling: { sampling_rate: 1.0 } }
       ],
       default_sampling_probability: 0.001,
       default_lower_bound_traces_per_second: 1.0 / (10.0 * 60.0)
@@ -119,6 +124,39 @@ Set `sampler` to
   )
 ```
 
+#### RemoteControlled sampler
+
+`RemoteControlled` sampler is a sampler that is controller by jaeger agent. It starts out with `Probabilistic` sampler. It polls the jaeger-agent and changes sampling strategy accordingly. Set `sampler` to `Jaeger::Client::Samplers::RemoteControlled.new(service_name: 'service_name')`.
+
+RemoteControlled sampler options:
+
+| Param             | Required | Description |
+|-------------------|----------|-------------|
+| service_name      | x | name of the current service / application, same as given to Tracer |
+| sampler           |   | initial sampler to use prior to retrieving strategies from Agent |
+| refresh_interval  |   | interval in seconds before sampling strategy refreshes (0 to not refresh, defaults to 60) |
+| host              |   | host for jaeger-agent (defaults to 'localhost') |
+| port              |   | port for jaeger-agent for SamplingManager endpoint (defaults to 5778) |
+| logger            |   | logger for communication between jaeger-agent (default to $stdout logger) |
+
+### TraceContext compatible header propagation
+
+It is possible to use [W3C Trace Context](https://www.w3.org/TR/trace-context/#overview) headers to propagate the tracing information.
+
+To set it up you need to change FORMAT_RACK injector and extractor.
+
+```ruby
+OpenTracing.global_tracer = Jaeger::Client.build(
+  service_name: 'service_name',
+  injectors: {
+    OpenTracing::FORMAT_RACK => [Jaeger::Injectors::TraceContextRackCodec]
+  },
+  extractors: {
+    OpenTracing::FORMAT_RACK => [Jaeger::Extractors::TraceContextRackCodec]
+  }
+)
+```
+
 ### Zipkin HTTP B3 compatible header propagation
 
 Jaeger Tracer supports Zipkin B3 Propagation HTTP headers, which are used by a lot of Zipkin tracers. This means that you can use Jaeger in conjunction with OpenZipkin tracers.
@@ -141,6 +179,20 @@ It's also possible to set up multiple injectors and extractors. Each injector wi
 
 If multiple extractors is used then the span context from the first match will be returned.
 
+### Process Tags
+
+Jaeger Tracer allows you to define process level tags. By default the tracer provides `jaeger.version`, `ip` and `hostname`. You may want to overwrite `ip` or `hostname` if the tracer cannot auto-detect them.
+
+```ruby
+OpenTracing.global_tracer = Jaeger::Client.build(
+  service_name: 'service_name',
+  tags: {
+    'hostname' => 'custom-hostname',
+    'custom_tag' => 'custom-tag-value'
+  }
+)
+```
+
 ## Development
 
 After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake spec` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.

data/crossdock/Dockerfile

@@ -1,4 +1,4 @@
-FROM ruby:2.5-alpine
+FROM ruby:2.7-alpine
 
 ENV APP_HOME /app
 
@@ -12,6 +12,8 @@ ADD Gemfile Gemfile.lock jaeger-client.gemspec $APP_HOME/
 ADD lib/jaeger/client/version.rb $APP_HOME/lib/jaeger/client/
 ADD crossdock/Gemfile crossdock/Gemfile.lock $APP_HOME/crossdock/
 
+RUN gem install bundler -v $(tail -n 1 $APP_HOME/Gemfile.lock | awk '{$1=$1};1')
+
 RUN apk add --no-cache --virtual .app-builddeps build-base \
   && cd $APP_HOME && bundle install \
   && cd $APP_HOME/crossdock && bundle install \

data/crossdock/Gemfile

@@ -3,4 +3,4 @@ source 'https://rubygems.org'
 gem 'jaeger-client', path: '../'
 gem 'rack'
 gem 'sinatra'
-gem 'webrick', '~> 1.4.2'
+gem 'webrick', '~> 1.7'

data/crossdock/Gemfile.lock

@@ -1,26 +1,28 @@
 PATH
   remote: ..
   specs:
-    jaeger-client (0.7.1)
+    jaeger-client (1.2.0)
       opentracing (~> 0.3)
       thrift
 
 GEM
   remote: https://rubygems.org/
   specs:
-    mustermann (1.0.3)
-    opentracing (0.4.3)
-    rack (2.0.6)
-    rack-protection (2.0.4)
+    mustermann (1.1.1)
+      ruby2_keywords (~> 0.0.1)
+    opentracing (0.5.0)
+    rack (2.2.3)
+    rack-protection (2.1.0)
       rack
-    sinatra (2.0.4)
+    ruby2_keywords (0.0.5)
+    sinatra (2.1.0)
       mustermann (~> 1.0)
-      rack (~> 2.0)
-      rack-protection (= 2.0.4)
+      rack (~> 2.2)
+      rack-protection (= 2.1.0)
       tilt (~> 2.0)
-    thrift (0.11.0.0)
-    tilt (2.0.9)
-    webrick (1.4.2)
+    thrift (0.15.0)
+    tilt (2.0.10)
+    webrick (1.7.0)
 
 PLATFORMS
   ruby
@@ -29,7 +31,7 @@ DEPENDENCIES
   jaeger-client!
   rack
   sinatra
-  webrick (~> 1.4.2)
+  webrick (~> 1.7)
 
 BUNDLED WITH
-   1.16.2
+   2.3.5

data/crossdock/jaeger-docker-compose.yml

@@ -3,46 +3,51 @@ version: '2'
 services:
     jaeger-collector:
       image: jaegertracing/jaeger-collector
-      command: ["--cassandra.keyspace=jaeger_v1_dc1", "--cassandra.servers=cassandra", "--collector.zipkin.http-port=9411"]
+      command: ["--es.num-shards=1", "--es.num-replicas=0", "--es.server-urls=http://elasticsearch:9200", "--collector.zipkin.host-port=:9411"]
       ports:
         - "14269"
         - "14268:14268"
-        - "14267"
         - "14250"
         - "9411:9411"
+      environment:
+        - SPAN_STORAGE_TYPE=elasticsearch
+        - LOG_LEVEL=debug
       restart: on-failure
       depends_on:
-        - cassandra-schema
+        - elasticsearch
 
     jaeger-query:
       image: jaegertracing/jaeger-query
-      command: ["--cassandra.keyspace=jaeger_v1_dc1", "--cassandra.servers=cassandra"]
+      command: ["--es.num-shards=1", "--es.num-replicas=0", "--es.server-urls=http://elasticsearch:9200"]
       ports:
         - "16686:16686"
         - "16687"
+      environment:
+        - SPAN_STORAGE_TYPE=elasticsearch
+        - LOG_LEVEL=debug
       restart: on-failure
       depends_on:
-        - cassandra-schema
+        - elasticsearch
 
     jaeger-agent:
       image: jaegertracing/jaeger-agent
-      # FIXME temporarily switching back to tchannel
-      # https://github.com/jaegertracing/jaeger/issues/1229
-      command: ["--reporter.tchannel.host-port=jaeger-collector:14267"]
+      command: ["--reporter.grpc.host-port=jaeger-collector:14250", "--reporter.grpc.retry.max=1000"]
       ports:
         - "5775:5775/udp"
         - "6831:6831/udp"
         - "6832:6832/udp"
         - "5778:5778"
+      environment:
+        - LOG_LEVEL=debug
       restart: on-failure
       depends_on:
         - jaeger-collector
 
-    cassandra:
-      image: cassandra:3.9
+    elasticsearch:
+      image: docker.elastic.co/elasticsearch/elasticsearch-oss:6.8.3
+      environment:
+        - discovery.type=single-node
+      ports:
+        - "9200:9200/tcp"
 
-    cassandra-schema:
-      image: jaegertracing/jaeger-cassandra-schema
-      depends_on:
-        - cassandra
 

data/crossdock/server

@@ -36,9 +36,10 @@ class HttpServer < Sinatra::Application
       transport = downstream['transport']
 
       response[:downstream] =
-        if transport == 'HTTP'
+        case transport
+        when 'HTTP'
           call_downstream_http(downstream, server_span)
-        elsif transport == 'DUMMY'
+        when 'DUMMY'
           { notImplementedError: 'Dummy has not been implemented' }
         else
           { notImplementedError: "Unrecognized transport received: #{transport}" }
@@ -69,9 +70,10 @@ class HttpServer < Sinatra::Application
       transport = downstream['transport']
 
       response[:downstream] =
-        if transport == 'HTTP'
+        case transport
+        when 'HTTP'
           call_downstream_http(downstream, server_span)
-        elsif transport == 'DUMMY'
+        when 'DUMMY'
           { notImplementedError: 'Dummy has not been implemented' }
         else
           { notImplementedError: "Unrecognized transport received: #{transport}" }

data/jaeger-client.gemspec

@@ -8,6 +8,7 @@ Gem::Specification.new do |spec|
   spec.version       = Jaeger::Client::VERSION
   spec.authors       = ['SaleMove TechMovers']
   spec.email         = ['techmovers@salemove.com']
+  spec.metadata['rubygems_mfa_required'] = 'true'
 
   spec.summary       = 'OpenTracing Tracer implementation for Jaeger in Ruby'
   spec.description   = ''
@@ -20,12 +21,16 @@ Gem::Specification.new do |spec|
   spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
   spec.require_paths = ['lib']
 
-  spec.add_development_dependency 'bundler', '~> 1.14'
-  spec.add_development_dependency 'rake', '~> 10.0'
-  spec.add_development_dependency 'rspec', '~> 3.0'
-  spec.add_development_dependency 'rubocop', '~> 0.54.0'
-  spec.add_development_dependency 'rubocop-rspec', '~> 1.24.0'
+  spec.required_ruby_version = ['>= 2.7', '< 3.2']
+
+  spec.add_development_dependency 'bundler'
+  spec.add_development_dependency 'rake', '~> 13.0'
+  spec.add_development_dependency 'rspec', '~> 3.10'
+  spec.add_development_dependency 'rubocop', '~> 1.25'
+  spec.add_development_dependency 'rubocop-rake', '~> 0.6'
+  spec.add_development_dependency 'rubocop-rspec', '~> 2.8'
   spec.add_development_dependency 'timecop', '~> 0.9'
+  spec.add_development_dependency 'webmock', '~> 3.14'
 
   spec.add_dependency 'opentracing', '~> 0.3'
   spec.add_dependency 'thrift'

data/lib/jaeger/client/version.rb

@@ -2,6 +2,6 @@
 
 module Jaeger
   module Client
-    VERSION = '0.10.0'.freeze
+    VERSION = '1.3.0'
   end
 end

data/lib/jaeger/client.rb

@@ -1,10 +1,14 @@
 # frozen_string_literal: true
 
-$LOAD_PATH.push(File.dirname(__FILE__) + '/../../thrift/gen-rb')
+$LOAD_PATH.push("#{File.dirname(__FILE__)}/../../thrift/gen-rb")
 
 require 'opentracing'
 require 'jaeger/thrift/agent'
 require 'logger'
+require 'time'
+require 'net/http'
+require 'cgi'
+require 'json'
 
 require_relative 'tracer'
 require_relative 'span'
@@ -21,6 +25,8 @@ require_relative 'encoders/thrift_encoder'
 require_relative 'injectors'
 require_relative 'extractors'
 require_relative 'rate_limiter'
+require_relative 'thrift_tag_builder'
+require_relative 'recurring_executor'
 
 module Jaeger
   module Client
@@ -37,21 +43,22 @@ module Jaeger
 
     DEFAULT_FLUSH_INTERVAL = 10
 
-    def self.build(host: '127.0.0.1',
+    def self.build(service_name:,
+                   host: '127.0.0.1',
                    port: 6831,
-                   service_name:,
                    flush_interval: DEFAULT_FLUSH_INTERVAL,
                    sampler: Samplers::Const.new(true),
-                   logger: Logger.new(STDOUT),
+                   logger: Logger.new($stdout),
                    sender: nil,
                    reporter: nil,
                    injectors: {},
-                   extractors: {})
-      encoder = Encoders::ThriftEncoder.new(service_name: service_name)
+                   extractors: {},
+                   tags: {})
+      encoder = Encoders::ThriftEncoder.new(service_name: service_name, tags: tags, logger: logger)
 
       if sender
         warn '[DEPRECATION] Passing `sender` directly to Jaeger::Client.build is deprecated.' \
-          'Please use `reporter` instead.'
+             'Please use `reporter` instead.'
       end
 
       reporter ||= Reporters::RemoteReporter.new(

data/lib/jaeger/encoders/thrift_encoder.rb

@@ -3,49 +3,59 @@
 module Jaeger
   module Encoders
     class ThriftEncoder
-      def initialize(service_name:)
+      def initialize(service_name:, logger:, tags: {})
         @service_name = service_name
-        @tags = [
-          Jaeger::Thrift::Tag.new(
-            'key' => 'jaeger.version',
-            'vType' => Jaeger::Thrift::TagType::STRING,
-            'vStr' => 'Ruby-' + Jaeger::Client::VERSION
-          ),
-          Jaeger::Thrift::Tag.new(
-            'key' => 'hostname',
-            'vType' => Jaeger::Thrift::TagType::STRING,
-            'vStr' => Socket.gethostname
-          )
-        ]
-        ipv4 = Socket.ip_address_list.find { |ai| ai.ipv4? && !ai.ipv4_loopback? }
-        unless ipv4.nil? # rubocop:disable Style/GuardClause
-          @tags << Jaeger::Thrift::Tag.new(
-            'key' => 'ip',
-            'vType' => Jaeger::Thrift::TagType::STRING,
-            'vStr' => ipv4.ip_address
-          )
-        end
+        @logger = logger
+        @tags = prepare_tags(tags)
+        @process = Jaeger::Thrift::Process.new('serviceName' => @service_name, 'tags' => @tags)
       end
 
       def encode(spans)
-        Jaeger::Thrift::Batch.new(
-          'process' => Jaeger::Thrift::Process.new(
-            'serviceName' => @service_name,
-            'tags' => @tags
-          ),
-          'spans' => spans.map(&method(:encode_span))
-        )
+        encode_batch(spans.map(&method(:encode_span)))
+      end
+
+      def encode_limited_size(spans, protocol_class, max_message_length)
+        batches = []
+        current_batch = []
+        current_batch_size = 0
+
+        max_spans_length = calculate_max_spans_length(protocol_class, max_message_length)
+
+        spans.each do |span|
+          encoded_span = encode_span(span)
+          span_size = message_size(encoded_span, protocol_class)
+
+          if span_size > max_spans_length
+            @logger.warn("Skip span #{span.operation_name} with size #{span_size}")
+            next
+          end
+
+          if (current_batch_size + span_size) > max_spans_length
+            batches << encode_batch(current_batch)
+            current_batch = []
+            current_batch_size = 0
+          end
+
+          current_batch << encoded_span
+          current_batch_size += span_size
+        end
+        batches << encode_batch(current_batch) unless current_batch.empty?
+        batches
       end
 
       private
 
+      def encode_batch(encoded_spans)
+        Jaeger::Thrift::Batch.new('process' => @process, 'spans' => encoded_spans)
+      end
+
       def encode_span(span)
         context = span.context
         start_ts, duration = build_timestamps(span)
 
         Jaeger::Thrift::Span.new(
-          'traceIdLow' => TraceId.uint64_id_to_int64(context.trace_id),
-          'traceIdHigh' => 0,
+          'traceIdLow' => TraceId.uint64_id_to_int64(trace_id_to_low(context.trace_id)),
+          'traceIdHigh' => TraceId.uint64_id_to_int64(trace_id_to_high(context.trace_id)),
           'spanId' => TraceId.uint64_id_to_int64(context.span_id),
           'parentSpanId' => TraceId.uint64_id_to_int64(context.parent_id),
           'operationName' => span.operation_name,
@@ -62,8 +72,8 @@ module Jaeger
         references.map do |ref|
           Jaeger::Thrift::SpanRef.new(
             'refType' => span_ref_type(ref.type),
-            'traceIdLow' => TraceId.uint64_id_to_int64(ref.context.trace_id),
-            'traceIdHigh' => 0,
+            'traceIdLow' => TraceId.uint64_id_to_int64(trace_id_to_low(ref.context.trace_id)),
+            'traceIdHigh' => TraceId.uint64_id_to_int64(trace_id_to_high(ref.context.trace_id)),
             'spanId' => TraceId.uint64_id_to_int64(ref.context.span_id)
           )
         end
@@ -87,6 +97,77 @@ module Jaeger
           nil
         end
       end
+
+      def prepare_tags(tags)
+        with_default_tags = tags.dup
+        with_default_tags['jaeger.version'] = "Ruby-#{Jaeger::Client::VERSION}"
+        with_default_tags['hostname'] ||= Socket.gethostname
+
+        unless with_default_tags['ip']
+          ipv4 = Socket.ip_address_list.find { |ai| ai.ipv4? && !ai.ipv4_loopback? }
+          with_default_tags['ip'] = ipv4.ip_address unless ipv4.nil?
+        end
+
+        with_default_tags.map do |key, value|
+          ThriftTagBuilder.build(key, value)
+        end
+      end
+
+      # Returns the right most 64 bits of trace id
+      def trace_id_to_low(trace_id)
+        trace_id & TraceId::MAX_64BIT_UNSIGNED_INT
+      end
+
+      # Returns the left most 64 bits of trace id
+      def trace_id_to_high(trace_id)
+        (trace_id >> 64) & TraceId::MAX_64BIT_UNSIGNED_INT
+      end
+
+      class DummyTransport
+        attr_accessor :size
+
+        def initialize
+          @size = 0
+        end
+
+        def write(buf)
+          @size += buf.size
+        end
+
+        def flush
+          @size = 0
+        end
+
+        def close; end
+      end
+
+      def message_size(message, protocol_class)
+        transport = DummyTransport.new
+        protocol = protocol_class.new(transport)
+        message.write(protocol)
+        transport.size
+      end
+
+      # Compact protocol have dynamic size of list header
+      # https://erikvanoosten.github.io/thrift-missing-specification/#_list_and_set_2
+      BATCH_SPANS_SIZE_WINDOW = 4
+
+      def empty_batch_size_cache
+        @empty_batch_size_cache ||= {}
+      end
+
+      def caclulate_empty_batch_size(protocol_class)
+        empty_batch_size_cache[protocol_class] ||=
+          message_size(encode_batch([]), protocol_class) + BATCH_SPANS_SIZE_WINDOW
+      end
+
+      def calculate_max_spans_length(protocol_class, max_message_length)
+        empty_batch_size = caclulate_empty_batch_size(protocol_class)
+        max_spans_length = max_message_length - empty_batch_size
+        return max_spans_length if max_spans_length.positive?
+
+        raise "Batch header have size #{empty_batch_size}, but limit #{max_message_length}"
+      end
     end
   end
 end