prometheus_exporter 2.0.8 → 2.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: af71a5656b5fe6a90de6fe6ad1b76efa05ccfb67d4fdb3cdcc740f539097546f
-  data.tar.gz: cd6ef704ecdc7110d5f78dd999354e1ea093adcba51f2355a0211e99e0fb3c67
+  metadata.gz: '042907307e8e1555111f34fb69d6cf66127e7742a76130952e3708a742ada24a'
+  data.tar.gz: 1e474618e9ddfa870a47e3ba7d7e55935677f3770d73968bb60457474b49502d
 SHA512:
-  metadata.gz: adf3c09a5609699bd055094bc509e997e320a72833b7c2beca03186d0c3d3fa0fa65fc005708b7faf496a9bf32c68c489cc7a1985525eed7a35ceb6b2d53a9df
-  data.tar.gz: 3975585eb3842c5cf409793d9257a184e33af062d4dd45e5bf64164bbba88c87d14aaea946bed519f9db1034afb8695fbdeecef337286bcaf00132cd2cd7a5f2
+  metadata.gz: '0137930945e0b254dbf0d7245d696d782044bfee9dd255ff022ffadaddafb47cdefc1aa6f83927bc84866d87fee761027590c2f08fbe730496e7099e68a883b7'
+  data.tar.gz: 562691a91a7f4a0e380b92236f3295177481763393db7b0db9d8763f6ac631d317e6bd21bd3368471029742b8e649a1c2a3884fa2aafe93d280aa353909d9ecd

data/.github/workflows/ci.yml

@@ -8,6 +8,13 @@ on:
   schedule:
     - cron: "0 0 * * 0" # weekly
 
+permissions:
+  contents: write
+  packages: write
+
+env:
+  DOCKER_REPO: ghcr.io/discourse/prometheus_exporter
+
 jobs:
   build:
     runs-on: ubuntu-latest
@@ -20,8 +27,8 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        ruby: ['2.6', '2.7', '3.0', '3.1', '3.2']
-        activerecord: [60, 61]
+        ruby: ['3.0', '3.1', '3.2', '3.3']
+        activerecord: [60, 61, 70]
 
     steps:
       - uses: actions/checkout@v2
@@ -42,12 +49,51 @@ jobs:
     needs: build
     runs-on: ubuntu-latest
 
+    outputs:
+      new_version_published: ${{ steps.release.outputs.new_version_published }}
+
     steps:
       - uses: actions/checkout@v2
 
       - name: Release gem
+        id: release
         uses: discourse/publish-rubygems-action@v2
         env:
           RUBYGEMS_API_KEY: ${{ secrets.RUBYGEMS_API_KEY }}
           GIT_EMAIL: team@discourse.org
           GIT_NAME: discoursebot
+
+  publish_docker:
+    needs: publish
+    if: needs.publish.outputs.new_version_published == 'true'
+    runs-on: ubuntu-latest
+    timeout-minutes: 20
+
+    steps:
+      - uses: actions/checkout@v3
+      - uses: docker/setup-qemu-action@v2
+      - uses: docker/setup-buildx-action@v2
+
+      - name: Set vars
+        id: vars
+        run: |
+          ruby -r ./lib/prometheus_exporter/version.rb -e 'print "version=#{PrometheusExporter::VERSION}"' >> $GITHUB_OUTPUT
+
+      - name: Login to Github Container Registry
+        uses: docker/login-action@v2
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Build and push images
+        uses: docker/build-push-action@v3
+        with:
+          context: .
+          push: true
+          platforms: linux/amd64,linux/arm64
+          build-args: |
+            GEM_VERSION=${{ steps.vars.outputs.version }}
+          tags: |
+            ${{ env.DOCKER_REPO }}:${{ steps.vars.outputs.version }}
+            ${{ env.DOCKER_REPO }}:latest

data/Appraisals

@@ -1,10 +1,14 @@
 # frozen_string_literal: true
 
 appraise "ar-60" do
-  # we are using this version as default in gemspec
-  # gem "activerecord", "~> 6.0.0"
+  gem "activerecord", "~> 6.0.0"
 end
 
 appraise "ar-61" do
   gem "activerecord", "~> 6.1.1"
 end
+
+appraise "ar-70" do
+  # latest version
+  gem "activerecord", "~> 7.1.2"
+end

data/CHANGELOG

@@ -1,3 +1,9 @@
+2.1.0 - 2024-08-01
+
+- FEATURE: good_job instrumentation
+- PERF: improve performance of histogram
+- DEV: use new metric collector pattern so we reuse code between collectors
+
 2.0.8 - 2023-01-20
 
 - FEATURE: attempting to make our first docker release

data/README.md

@@ -21,6 +21,7 @@ To learn more see [Instrumenting Rails with Prometheus](https://samsaffron.com/a
   * [Puma metrics](#puma-metrics)
   * [Unicorn metrics](#unicorn-process-metrics)
   * [Resque metrics](#resque-metrics)
+  * [GoodJob metrics](#goodjob-metrics)
   * [Custom type collectors](#custom-type-collectors)
   * [Multi process mode with custom collector](#multi-process-mode-with-custom-collector)
   * [GraphQL support](#graphql-support)
@@ -28,7 +29,7 @@ To learn more see [Instrumenting Rails with Prometheus](https://samsaffron.com/a
   * [Client default labels](#client-default-labels)
   * [Client default host](#client-default-host)
   * [Histogram mode](#histogram-mode)
-  * [Histogram - custom buckets](#histogram-custom-buckets)
+  * [Histogram - custom buckets](#histogram---custom-buckets)
 * [Transport concerns](#transport-concerns)
 * [JSON generation and parsing](#json-generation-and-parsing)
 * [Logging](#logging)
@@ -188,7 +189,7 @@ gem 'prometheus_exporter'
 In an initializer:
 
 ```ruby
-unless Rails.env == "test"
+unless Rails.env.test?
   require 'prometheus_exporter/middleware'
 
   # This reports stats per request like HTTP status and timings
@@ -341,7 +342,7 @@ You may also be interested in per-process stats. This collects memory and GC sta
 
 ```ruby
 # in an initializer
-unless Rails.env == "test"
+unless Rails.env.test?
   require 'prometheus_exporter/instrumentation'
 
   # this reports basic process stats like RSS and GC info
@@ -522,7 +523,7 @@ All metrics have labels for `job_name` and `queue_name`.
 In an initializer:
 
 ```ruby
-unless Rails.env == "test"
+unless Rails.env.test?
   require 'prometheus_exporter/instrumentation'
   PrometheusExporter::Instrumentation::DelayedJob.register_plugin
 end
@@ -533,6 +534,7 @@ end
 | Type    | Name                                      | Description                                                        | Labels     |
 | ---     | ---                                       | ---                                                                | ---        |
 | Counter | `delayed_job_duration_seconds`            | Total time spent in delayed jobs                                   | `job_name` |
+| Counter | `delayed_job_latency_seconds_total`       | Total delayed jobs latency                                         | `job_name` |
 | Counter | `delayed_jobs_total`                      | Total number of delayed jobs executed                              | `job_name` |
 | Gauge   | `delayed_jobs_enqueued`                   | Number of enqueued delayed jobs                                    | -          |
 | Gauge   | `delayed_jobs_pending`                    | Number of pending delayed jobs                                     | -          |
@@ -542,13 +544,14 @@ end
 | Summary | `delayed_job_attempts_summary`            | Summary of the amount of attempts it takes delayed jobs to succeed | -          |
 
 All metrics have labels for `job_name` and `queue_name`.
+`delayed_job_latency_seconds_total` is considering delayed job's [sleep_delay](https://github.com/collectiveidea/delayed_job#:~:text=If%20no%20jobs%20are%20found%2C%20the%20worker%20sleeps%20for%20the%20amount%20of%20time%20specified%20by%20the%20sleep%20delay%20option.%20Set%20Delayed%3A%3AWorker.sleep_delay%20%3D%2060%20for%20a%2060%20second%20sleep%20time.) parameter, so please be aware of this in case you are looking for high latency precision.
 
 #### Hutch Message Processing Tracer
 
 Capture [Hutch](https://github.com/gocardless/hutch) metrics (how many jobs ran? how many failed? how long did they take?)
 
 ```ruby
-unless Rails.env == "test"
+unless Rails.env.test?
   require 'prometheus_exporter/instrumentation'
   Hutch::Config.set(:tracer, PrometheusExporter::Instrumentation::Hutch)
 end
@@ -570,7 +573,7 @@ Request Queueing is defined as the time it takes for a request to reach your app
 
 As this metric starts before `prometheus_exporter` can handle the request, you must add a specific HTTP header as early in your infrastructure as possible (we recommend your load balancer or reverse proxy).
 
-The Amazon Application Load Balancer [request tracing header](https://docs.aws.amazon.com/elasticloadbalancing/latest/application/load-balancer-request-tracing.html) is natively supported. If you are using another upstream entrypoint, you may configure your HTTP server / load balancer to add a header `X-Request-Start: t=<MSEC>` when passing the request upstream. For more information, please consult your software manual.
+The Amazon Application Load Balancer [request tracing header](https://docs.aws.amazon.com/elasticloadbalancing/latest/application/load-balancer-request-tracing.html) is natively supported. If you are using another upstream entrypoint, you may configure your HTTP server / load balancer to add a header `X-Request-Start: t=<MSEC>` when passing the request upstream. Please keep in mind request time start is reported as epoch time (in seconds) and lacks precision, which may introduce additional latency in reported metrics. For more information, please consult your software manual.
 
 Hint: we aim to be API-compatible with the big APM solutions, so if you've got requests queueing time configured for them, it should be expected to also work with `prometheus_exporter`.
 
@@ -627,6 +630,29 @@ PrometheusExporter::Instrumentation::Resque.start
 | Gauge | `resque_workers`        | Total number of Resque workers running |
 | Gauge | `resque_working`        | Total number of Resque workers working |
 
+### GoodJob metrics
+
+The metrics are generated from the database using the relevant scopes. To start monitoring your GoodJob
+installation, you'll need to start the instrumentation:
+
+```ruby
+# e.g. config/initializers/good_job.rb
+require 'prometheus_exporter/instrumentation'
+PrometheusExporter::Instrumentation::GoodJob.start
+```
+
+#### Metrics collected by GoodJob Instrumentation
+
+| Type  | Name                 | Description                             |
+| ---   |----------------------|-----------------------------------------|
+| Gauge | `good_job_scheduled` | Total number of scheduled GoodJob jobs. |
+| Gauge | `good_job_retried`   | Total number of retried GoodJob jobs.   |
+| Gauge | `good_job_queued`    | Total number of queued GoodJob jobs.    |
+| Gauge | `good_job_running`   | Total number of running GoodJob jobs.   |
+| Gauge | `good_job_finished`  | Total number of finished GoodJob jobs.  |
+| Gauge | `good_job_succeeded` | Total number of succeeded GoodJob jobs. |
+| Gauge | `good_job_discarded` | Total number of discarded GoodJob jobs  |
+
 ### Unicorn process metrics
 
 In order to gather metrics from unicorn processes, we use `rainbows`, which exposes `Rainbows::Linux.tcp_listener_stats` to gather information about active workers and queued requests. To start monitoring your unicorn processes, you'll need to know both the path to unicorn PID file and the listen address (`pid_file` and `listen` in your unicorn config file)

data/gemfiles/ar_70.gemfile

@@ -0,0 +1,5 @@
+# This file was generated by Appraisal
+
+source "https://rubygems.org"
+
+gemspec path: "../"

data/lib/prometheus_exporter/client.rb

@@ -1,8 +1,7 @@
 # frozen_string_literal: true
 
-require 'socket'
-require 'thread'
-require 'logger'
+require "socket"
+require "logger"
 
 module PrometheusExporter
   class Client
@@ -25,7 +24,9 @@ module PrometheusExporter
           keys: keys,
           value: value
         }
-        values[:prometheus_exporter_action] = prometheus_exporter_action if prometheus_exporter_action
+        values[
+          :prometheus_exporter_action
+        ] = prometheus_exporter_action if prometheus_exporter_action
         values[:opts] = @opts if @opts
         values
       end
@@ -57,8 +58,11 @@ module PrometheusExporter
     attr_reader :logger
 
     def initialize(
-      host: ENV.fetch('PROMETHEUS_EXPORTER_HOST', 'localhost'),
-      port: ENV.fetch('PROMETHEUS_EXPORTER_PORT', PrometheusExporter::DEFAULT_PORT),
+      host: ENV.fetch("PROMETHEUS_EXPORTER_HOST", "localhost"),
+      port: ENV.fetch(
+        "PROMETHEUS_EXPORTER_PORT",
+        PrometheusExporter::DEFAULT_PORT
+      ),
       max_queue_size: nil,
       thread_sleep: 0.5,
       json_serializer: nil,
@@ -90,7 +94,8 @@ module PrometheusExporter
       @mutex = Mutex.new
       @thread_sleep = thread_sleep
 
-      @json_serializer = json_serializer == :oj ? PrometheusExporter::OjCompat : JSON
+      @json_serializer =
+        json_serializer == :oj ? PrometheusExporter::OjCompat : JSON
 
       @custom_labels = custom_labels
     end
@@ -100,7 +105,14 @@ module PrometheusExporter
     end
 
     def register(type, name, help, opts = nil)
-      metric = RemoteMetric.new(type: type, name: name, help: help, client: self, opts: opts)
+      metric =
+        RemoteMetric.new(
+          type: type,
+          name: name,
+          help: help,
+          client: self,
+          opts: opts
+        )
       @metrics << metric
       metric
     end
@@ -161,9 +173,7 @@ module PrometheusExporter
       @mutex.synchronize do
         wait_for_empty_queue_with_timeout(wait_timeout_seconds)
         @worker_thread&.kill
-        while @worker_thread&.alive?
-          sleep 0.001
-        end
+        sleep 0.001 while @worker_thread&.alive?
         @worker_thread = nil
         close_socket!
       end
@@ -183,12 +193,13 @@ module PrometheusExporter
         @mutex.synchronize do
           return if @worker_thread&.alive?
 
-          @worker_thread = Thread.new do
-            while true
-              worker_loop
-              sleep @thread_sleep
+          @worker_thread =
+            Thread.new do
+              while true
+                worker_loop
+                sleep @thread_sleep
+              end
             end
-          end
         end
       end
     rescue ThreadError => e
@@ -212,7 +223,8 @@ module PrometheusExporter
     end
 
     def close_socket_if_old!
-      if @socket_pid == Process.pid && @socket && @socket_started && ((@socket_started + MAX_SOCKET_AGE) < Time.now.to_f)
+      if @socket_pid == Process.pid && @socket && @socket_started &&
+           ((@socket_started + MAX_SOCKET_AGE) < Time.now.to_f)
         close_socket!
       end
     end
@@ -240,7 +252,7 @@ module PrometheusExporter
       end
 
       nil
-    rescue
+    rescue StandardError
       @socket = nil
       @socket_started = nil
       @socket_pid = nil
@@ -250,7 +262,10 @@ module PrometheusExporter
     def wait_for_empty_queue_with_timeout(timeout_seconds)
       start_time = ::Process.clock_gettime(::Process::CLOCK_MONOTONIC)
       while @queue.length > 0
-        break if start_time + timeout_seconds < ::Process.clock_gettime(::Process::CLOCK_MONOTONIC)
+        if start_time + timeout_seconds <
+             ::Process.clock_gettime(::Process::CLOCK_MONOTONIC)
+          break
+        end
         sleep(0.05)
       end
     end

data/lib/prometheus_exporter/instrumentation/delayed_job.rb

@@ -31,6 +31,7 @@ module PrometheusExporter::Instrumentation
     def call(job, max_attempts, enqueued_count, pending_count, *args, &block)
       success = false
       start = ::Process.clock_gettime(::Process::CLOCK_MONOTONIC)
+      latency = Time.current - job.run_at
       attempts = job.attempts + 1 # Increment because we're adding the current attempt
       result = block.call(job, *args)
       success = true
@@ -44,6 +45,7 @@ module PrometheusExporter::Instrumentation
         queue_name: job.queue,
         success: success,
         duration: duration,
+        latency: latency,
         attempts: attempts,
         max_attempts: max_attempts,
         enqueued: enqueued_count,

data/lib/prometheus_exporter/instrumentation/good_job.rb

@@ -0,0 +1,30 @@
+# frozen_string_literal: true
+
+# collects stats from GoodJob
+module PrometheusExporter::Instrumentation
+  class GoodJob < PeriodicStats
+    def self.start(client: nil, frequency: 30)
+      good_job_collector = new
+      client ||= PrometheusExporter::Client.default
+
+      worker_loop do
+        client.send_json(good_job_collector.collect)
+      end
+
+      super
+    end
+
+    def collect
+      {
+        type: "good_job",
+        scheduled: ::GoodJob::Job.scheduled.size,
+        retried: ::GoodJob::Job.retried.size,
+        queued: ::GoodJob::Job.queued.size,
+        running: ::GoodJob::Job.running.size,
+        finished: ::GoodJob::Job.finished.size,
+        succeeded: ::GoodJob::Job.succeeded.size,
+        discarded: ::GoodJob::Job.discarded.size
+      }
+    end
+  end
+end

data/lib/prometheus_exporter/instrumentation/sidekiq.rb

@@ -1,18 +1,19 @@
 # frozen_string_literal: true
 
-require 'yaml'
+require "yaml"
 
 module PrometheusExporter::Instrumentation
-  JOB_WRAPPER_CLASS_NAME = 'ActiveJob::QueueAdapters::SidekiqAdapter::JobWrapper'
-  DELAYED_CLASS_NAMES = [
-    'Sidekiq::Extensions::DelayedClass',
-    'Sidekiq::Extensions::DelayedModel',
-    'Sidekiq::Extensions::DelayedMailer',
+  JOB_WRAPPER_CLASS_NAME =
+    "ActiveJob::QueueAdapters::SidekiqAdapter::JobWrapper"
+  DELAYED_CLASS_NAMES = %w[
+    Sidekiq::Extensions::DelayedClass
+    Sidekiq::Extensions::DelayedModel
+    Sidekiq::Extensions::DelayedMailer
   ]
 
   class Sidekiq
     def self.death_handler
-      -> (job, ex) do
+      ->(job, ex) do
         job_is_fire_and_forget = job["retry"] == false
 
         worker_class = Object.const_get(job["class"])
@@ -43,7 +44,8 @@ module PrometheusExporter::Instrumentation
     end
 
     def initialize(options = { client: nil })
-      @client = options.fetch(:client, nil) || PrometheusExporter::Client.default
+      @client =
+        options.fetch(:client, nil) || PrometheusExporter::Client.default
     end
 
     def call(worker, msg, queue)
@@ -82,7 +84,7 @@ module PrometheusExporter::Instrumentation
     end
 
     def self.get_job_wrapper_name(msg)
-      msg['wrapped']
+      msg["wrapped"]
     end
 
     def self.get_delayed_name(msg, class_name)
@@ -90,17 +92,17 @@ module PrometheusExporter::Instrumentation
         # fallback to class_name since we're relying on the internal implementation
         # of the delayed extensions
         # https://github.com/mperham/sidekiq/blob/master/lib/sidekiq/extensions/class_methods.rb
-        (target, method_name, _args) = YAML.load(msg['args'].first) # rubocop:disable Security/YAMLLoad
+        target, method_name, _args = YAML.load(msg["args"].first)
         if target.class == Class
           "#{target.name}##{method_name}"
         else
           "#{target.class.name}##{method_name}"
         end
       rescue Psych::DisallowedClass, ArgumentError
-        parsed = Psych.parse(msg['args'].first)
+        parsed = Psych.parse(msg["args"].first)
         children = parsed.root.children
-        target = (children[0].value || children[0].tag).sub('!', '')
-        method_name = (children[1].value || children[1].tag).sub(':', '')
+        target = (children[0].value || children[0].tag).sub("!", "")
+        method_name = (children[1].value || children[1].tag).sub(":", "")
 
         if target && method_name
           "#{target}##{method_name}"
@@ -108,7 +110,7 @@ module PrometheusExporter::Instrumentation
           class_name
         end
       end
-    rescue
+    rescue StandardError
       class_name
     end
   end

data/lib/prometheus_exporter/instrumentation.rb

@@ -15,3 +15,4 @@ require_relative "instrumentation/unicorn"
 require_relative "instrumentation/active_record"
 require_relative "instrumentation/shoryuken"
 require_relative "instrumentation/resque"
+require_relative "instrumentation/good_job"

data/lib/prometheus_exporter/metric/histogram.rb

@@ -91,7 +91,7 @@ module PrometheusExporter::Metric
     end
 
     def fill_buckets(value, buckets)
-      @buckets.reverse.each do |b|
+      @buckets.reverse_each do |b|
         break if value > b
         buckets[b] += 1
       end

data/lib/prometheus_exporter/server/active_record_collector.rb

@@ -2,7 +2,8 @@
 
 module PrometheusExporter::Server
   class ActiveRecordCollector < TypeCollector
-    MAX_ACTIVERECORD_METRIC_AGE = 60
+    MAX_METRIC_AGE = 60
+
     ACTIVE_RECORD_GAUGES = {
       connections: "Total connections in pool",
       busy: "Connections in use in pool",
@@ -13,7 +14,12 @@ module PrometheusExporter::Server
     }
 
     def initialize
-      @active_record_metrics = []
+      @active_record_metrics = MetricsContainer.new(ttl: MAX_METRIC_AGE)
+      @active_record_metrics.filter = -> (new_metric, old_metric) do
+        new_metric["pid"] == old_metric["pid"] &&
+        new_metric["hostname"] == old_metric["hostname"] &&
+        new_metric["metric_labels"]["pool_name"] == old_metric["metric_labels"]["pool_name"]
+      end
     end
 
     def type
@@ -26,7 +32,7 @@ module PrometheusExporter::Server
       metrics = {}
 
       @active_record_metrics.map do |m|
-        metric_key = (m["metric_labels"] || {}).merge("pid" => m["pid"])
+        metric_key = (m["metric_labels"] || {}).merge("pid" => m["pid"], "hostname" => m["hostname"])
         metric_key.merge!(m["custom_labels"]) if m["custom_labels"]
 
         ACTIVE_RECORD_GAUGES.map do |k, help|
@@ -42,16 +48,6 @@ module PrometheusExporter::Server
     end
 
     def collect(obj)
-      now = ::Process.clock_gettime(::Process::CLOCK_MONOTONIC)
-
-      obj["created_at"] = now
-
-      @active_record_metrics.delete_if do |current|
-        (obj["pid"] == current["pid"] && obj["hostname"] == current["hostname"] &&
-         obj["metric_labels"]["pool_name"] == current["metric_labels"]["pool_name"]) ||
-          (current["created_at"] + MAX_ACTIVERECORD_METRIC_AGE < now)
-      end
-
       @active_record_metrics << obj
     end
   end

data/lib/prometheus_exporter/server/collector.rb

@@ -23,6 +23,7 @@ module PrometheusExporter::Server
       register_collector(ActiveRecordCollector.new)
       register_collector(ShoryukenCollector.new)
       register_collector(ResqueCollector.new)
+      register_collector(GoodJobCollector.new)
     end
 
     def register_collector(collector)

data/lib/prometheus_exporter/server/delayed_job_collector.rb

@@ -5,6 +5,7 @@ module PrometheusExporter::Server
     def initialize
       @delayed_jobs_total = nil
       @delayed_job_duration_seconds = nil
+      @delayed_job_latency_seconds_total = nil
       @delayed_jobs_total = nil
       @delayed_failed_jobs_total = nil
       @delayed_jobs_max_attempts_reached_total = nil
@@ -25,6 +26,7 @@ module PrometheusExporter::Server
 
       ensure_delayed_job_metrics
       @delayed_job_duration_seconds.observe(obj["duration"], counter_labels)
+      @delayed_job_latency_seconds_total.observe(obj["latency"], counter_labels)
       @delayed_jobs_total.observe(1, counter_labels)
       @delayed_failed_jobs_total.observe(1, counter_labels) if !obj["success"]
       @delayed_jobs_max_attempts_reached_total.observe(1, counter_labels) if obj["attempts"] >= obj["max_attempts"]
@@ -38,7 +40,7 @@ module PrometheusExporter::Server
 
     def metrics
       if @delayed_jobs_total
-        [@delayed_job_duration_seconds, @delayed_jobs_total, @delayed_failed_jobs_total,
+        [@delayed_job_duration_seconds, @delayed_job_latency_seconds_total, @delayed_jobs_total, @delayed_failed_jobs_total,
          @delayed_jobs_max_attempts_reached_total, @delayed_job_duration_seconds_summary, @delayed_job_attempts_summary,
          @delayed_jobs_enqueued, @delayed_jobs_pending]
       else
@@ -55,6 +57,10 @@ module PrometheusExporter::Server
         PrometheusExporter::Metric::Counter.new(
           "delayed_job_duration_seconds", "Total time spent in delayed jobs.")
 
+        @delayed_job_latency_seconds_total =
+        PrometheusExporter::Metric::Counter.new(
+          "delayed_job_latency_seconds_total", "Total delayed jobs latency.")
+
         @delayed_jobs_total =
         PrometheusExporter::Metric::Counter.new(
           "delayed_jobs_total", "Total number of delayed jobs executed.")

data/lib/prometheus_exporter/server/good_job_collector.rb

@@ -0,0 +1,52 @@
+# frozen_string_literal: true
+
+module PrometheusExporter::Server
+  class GoodJobCollector < TypeCollector
+    MAX_METRIC_AGE = 30
+    GOOD_JOB_GAUGES = {
+      scheduled: "Total number of scheduled GoodJob jobs.",
+      retried: "Total number of retried GoodJob jobs.",
+      queued: "Total number of queued GoodJob jobs.",
+      running: "Total number of running GoodJob jobs.",
+      finished: "Total number of finished GoodJob jobs.",
+      succeeded: "Total number of succeeded GoodJob jobs.",
+      discarded: "Total number of discarded GoodJob jobs."
+    }
+
+    def initialize
+      @good_job_metrics = MetricsContainer.new(ttl: MAX_METRIC_AGE)
+      @gauges = {}
+    end
+
+    def type
+      "good_job"
+    end
+
+    def metrics
+      return [] if good_job_metrics.length == 0
+
+      good_job_metrics.map do |metric|
+        labels = metric.fetch("custom_labels", {})
+
+        GOOD_JOB_GAUGES.map do |name, help|
+          value = metric[name.to_s]
+
+          if value
+            gauge = gauges[name] ||= PrometheusExporter::Metric::Gauge.new("good_job_#{name}", help)
+            gauge.observe(value, labels)
+          end
+        end
+      end
+
+      gauges.values
+    end
+
+    def collect(object)
+      @good_job_metrics << object
+    end
+
+    private
+
+    attr_reader :good_job_metrics, :gauges
+  end
+end

data/lib/prometheus_exporter/server/process_collector.rb

@@ -3,7 +3,8 @@
 module PrometheusExporter::Server
 
   class ProcessCollector < TypeCollector
-    MAX_PROCESS_METRIC_AGE = 60
+    MAX_METRIC_AGE = 60
+
     PROCESS_GAUGES = {
       heap_free_slots: "Free ruby heap slots.",
       heap_live_slots: "Used ruby heap slots.",
@@ -21,7 +22,10 @@ module PrometheusExporter::Server
     }
 
     def initialize
-      @process_metrics = []
+      @process_metrics = MetricsContainer.new(ttl: MAX_METRIC_AGE)
+      @process_metrics.filter = -> (new_metric, old_metric) do
+        new_metric["pid"] == old_metric["pid"] && new_metric["hostname"] == old_metric["hostname"]
+      end
     end
 
     def type
@@ -34,8 +38,8 @@ module PrometheusExporter::Server
       metrics = {}
 
       @process_metrics.map do |m|
-        metric_key = m["metric_labels"].merge("pid" => m["pid"])
-        metric_key.merge!(m["custom_labels"] || {})
+        metric_key = (m["metric_labels"] || {}).merge("pid" => m["pid"], "hostname" => m["hostname"])
+        metric_key.merge!(m["custom_labels"]) if m["custom_labels"]
 
         PROCESS_GAUGES.map do |k, help|
           k = k.to_s
@@ -58,15 +62,6 @@ module PrometheusExporter::Server
     end
 
     def collect(obj)
-      now = ::Process.clock_gettime(::Process::CLOCK_MONOTONIC)
-
-      obj["created_at"] = now
-
-      @process_metrics.delete_if do |current|
-        (obj["pid"] == current["pid"] && obj["hostname"] == current["hostname"]) ||
-          (current["created_at"] + MAX_PROCESS_METRIC_AGE < now)
-      end
-
       @process_metrics << obj
     end
   end

data/lib/prometheus_exporter/server/puma_collector.rb

@@ -14,7 +14,7 @@ module PrometheusExporter::Server
     }
 
     def initialize
-      @puma_metrics = MetricsContainer.new
+      @puma_metrics = MetricsContainer.new(ttl: MAX_PUMA_METRIC_AGE)
       @puma_metrics.filter = -> (new_metric, old_metric) do
         new_metric["pid"] == old_metric["pid"] && new_metric["hostname"] == old_metric["hostname"]
       end

data/lib/prometheus_exporter/server/resque_collector.rb

@@ -2,7 +2,7 @@
 
 module PrometheusExporter::Server
   class ResqueCollector < TypeCollector
-    MAX_RESQUE_METRIC_AGE = 30
+    MAX_METRIC_AGE = 30
     RESQUE_GAUGES = {
       processed_jobs: "Total number of processed Resque jobs.",
       failed_jobs: "Total number of failed Resque jobs.",
@@ -13,7 +13,7 @@ module PrometheusExporter::Server
     }
 
     def initialize
-      @resque_metrics = []
+      @resque_metrics = MetricsContainer.new(ttl: MAX_METRIC_AGE)
       @gauges = {}
     end
 
@@ -40,11 +40,7 @@ module PrometheusExporter::Server
     end
 
     def collect(object)
-      now = ::Process.clock_gettime(::Process::CLOCK_MONOTONIC)
-
-      object["created_at"] = now
-      resque_metrics.delete_if { |metric| metric["created_at"] + MAX_RESQUE_METRIC_AGE < now }
-      resque_metrics << object
+      @resque_metrics << object
     end
 
     private

data/lib/prometheus_exporter/server/sidekiq_process_collector.rb

@@ -2,7 +2,7 @@
 
 module PrometheusExporter::Server
   class SidekiqProcessCollector < PrometheusExporter::Server::TypeCollector
-    MAX_SIDEKIQ_METRIC_AGE = 60
+    MAX_METRIC_AGE = 60
 
     SIDEKIQ_PROCESS_GAUGES = {
       'busy' => 'Number of running jobs',
@@ -12,7 +12,7 @@ module PrometheusExporter::Server
     attr_reader :sidekiq_metrics, :gauges
 
     def initialize
-      @sidekiq_metrics = MetricsContainer.new(ttl: MAX_SIDEKIQ_METRIC_AGE)
+      @sidekiq_metrics = MetricsContainer.new(ttl: MAX_METRIC_AGE)
       @gauges = {}
     end
 

data/lib/prometheus_exporter/server/sidekiq_queue_collector.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 module PrometheusExporter::Server
   class SidekiqQueueCollector < TypeCollector
-    MAX_SIDEKIQ_METRIC_AGE = 60
+    MAX_METRIC_AGE = 60
 
     SIDEKIQ_QUEUE_GAUGES = {
       'backlog' => 'Size of the sidekiq queue.',
@@ -11,7 +11,7 @@ module PrometheusExporter::Server
     attr_reader :sidekiq_metrics, :gauges
 
     def initialize
-      @sidekiq_metrics = MetricsContainer.new
+      @sidekiq_metrics = MetricsContainer.new(ttl: MAX_METRIC_AGE)
       @gauges = {}
     end
 

data/lib/prometheus_exporter/server/sidekiq_stats_collector.rb

@@ -2,7 +2,7 @@
 
 module PrometheusExporter::Server
   class SidekiqStatsCollector < TypeCollector
-    MAX_SIDEKIQ_METRIC_AGE = 60
+    MAX_METRIC_AGE = 60
 
     SIDEKIQ_STATS_GAUGES = {
       'dead_size' => 'Size of dead the queue',
@@ -18,7 +18,7 @@ module PrometheusExporter::Server
     attr_reader :sidekiq_metrics, :gauges
 
     def initialize
-      @sidekiq_metrics = MetricsContainer.new(ttl: MAX_SIDEKIQ_METRIC_AGE)
+      @sidekiq_metrics = MetricsContainer.new(ttl: MAX_METRIC_AGE)
       @gauges = {}
     end
 

data/lib/prometheus_exporter/server/unicorn_collector.rb

@@ -2,47 +2,46 @@
 
 # custom type collector for prometheus_exporter for handling the metrics sent from
 # PrometheusExporter::Instrumentation::Unicorn
-class PrometheusExporter::Server::UnicornCollector < PrometheusExporter::Server::TypeCollector
-  MAX_UNICORN_METRIC_AGE = 60
-
-  UNICORN_GAUGES = {
-    workers: 'Number of unicorn workers.',
-    active_workers: 'Number of active unicorn workers',
-    request_backlog: 'Number of requests waiting to be processed by a unicorn worker.'
-  }.freeze
-
-  def initialize
-    @unicorn_metrics = []
-  end
+module PrometheusExporter::Server
+  class UnicornCollector < PrometheusExporter::Server::TypeCollector
+    MAX_METRIC_AGE = 60
+
+    UNICORN_GAUGES = {
+      workers: 'Number of unicorn workers.',
+      active_workers: 'Number of active unicorn workers',
+      request_backlog: 'Number of requests waiting to be processed by a unicorn worker.'
+    }.freeze
+
+    def initialize
+      @unicorn_metrics = MetricsContainer.new(ttl: MAX_METRIC_AGE)
+    end
 
-  def type
-    'unicorn'
-  end
+    def type
+      'unicorn'
+    end
 
-  def metrics
-    return [] if @unicorn_metrics.length.zero?
+    def metrics
+      return [] if @unicorn_metrics.length.zero?
 
-    metrics = {}
+      metrics = {}
 
-    @unicorn_metrics.map do |m|
-      labels = m["custom_labels"] || {}
+      @unicorn_metrics.map do |m|
+        labels = m["custom_labels"] || {}
 
-      UNICORN_GAUGES.map do |k, help|
-        k = k.to_s
-        if (v = m[k])
-          g = metrics[k] ||= PrometheusExporter::Metric::Gauge.new("unicorn_#{k}", help)
-          g.observe(v, labels)
+        UNICORN_GAUGES.map do |k, help|
+          k = k.to_s
+          if (v = m[k])
+            g = metrics[k] ||= PrometheusExporter::Metric::Gauge.new("unicorn_#{k}", help)
+            g.observe(v, labels)
+          end
         end
       end
-    end
 
-    metrics.values
-  end
+      metrics.values
+    end
 
-  def collect(obj)
-    now = ::Process.clock_gettime(::Process::CLOCK_MONOTONIC)
-    obj["created_at"] = now
-    @unicorn_metrics.delete_if { |m| m['created_at'] + MAX_UNICORN_METRIC_AGE < now }
-    @unicorn_metrics << obj
+    def collect(obj)
+      @unicorn_metrics << obj
+    end
   end
 end

data/lib/prometheus_exporter/server/web_server.rb

@@ -1,9 +1,9 @@
 # frozen_string_literal: true
 
-require 'webrick'
-require 'timeout'
-require 'zlib'
-require 'stringio'
+require "webrick"
+require "timeout"
+require "zlib"
+require "stringio"
 
 module PrometheusExporter::Server
   class WebServer
@@ -18,11 +18,23 @@ module PrometheusExporter::Server
       @auth = opts[:auth]
       @realm = opts[:realm] || PrometheusExporter::DEFAULT_REALM
 
-      @metrics_total = PrometheusExporter::Metric::Counter.new("collector_metrics_total", "Total metrics processed by exporter web.")
+      @metrics_total =
+        PrometheusExporter::Metric::Counter.new(
+          "collector_metrics_total",
+          "Total metrics processed by exporter web."
+        )
 
-      @sessions_total = PrometheusExporter::Metric::Counter.new("collector_sessions_total", "Total send_metric sessions processed by exporter web.")
+      @sessions_total =
+        PrometheusExporter::Metric::Counter.new(
+          "collector_sessions_total",
+          "Total send_metric sessions processed by exporter web."
+        )
 
-      @bad_metrics_total = PrometheusExporter::Metric::Counter.new("collector_bad_metrics_total", "Total mis-handled metrics by collector.")
+      @bad_metrics_total =
+        PrometheusExporter::Metric::Counter.new(
+          "collector_bad_metrics_total",
+          "Total mis-handled metrics by collector."
+        )
 
       @metrics_total.observe(0)
       @sessions_total.observe(0)
@@ -34,7 +46,7 @@ module PrometheusExporter::Server
       if @verbose
         @access_log = [
           [$stderr, WEBrick::AccessLog::COMMON_LOG_FORMAT],
-          [$stderr, WEBrick::AccessLog::REFERER_LOG_FORMAT],
+          [$stderr, WEBrick::AccessLog::REFERER_LOG_FORMAT]
         ]
         @logger = WEBrick::Log.new(log_target || $stderr)
       else
@@ -42,23 +54,26 @@ module PrometheusExporter::Server
         @logger = WEBrick::Log.new(log_target || "/dev/null")
       end
 
-      @logger.info "Using Basic Authentication via #{@auth}" if @verbose && @auth
+      if @verbose && @auth
+        @logger.info "Using Basic Authentication via #{@auth}"
+      end
 
-      if %w(ALL ANY).include?(@bind)
+      if %w[ALL ANY].include?(@bind)
         @logger.info "Listening on both 0.0.0.0/:: network interfaces"
         @bind = nil
       end
 
-      @server = WEBrick::HTTPServer.new(
-        Port: @port,
-        BindAddress: @bind,
-        Logger: @logger,
-        AccessLog: @access_log,
-      )
-
-      @server.mount_proc '/' do |req, res|
-        res['Content-Type'] = 'text/plain; charset=utf-8'
-        if req.path == '/metrics'
+      @server =
+        WEBrick::HTTPServer.new(
+          Port: @port,
+          BindAddress: @bind,
+          Logger: @logger,
+          AccessLog: @access_log
+        )
+
+      @server.mount_proc "/" do |req, res|
+        res["Content-Type"] = "text/plain; charset=utf-8"
+        if req.path == "/metrics"
           authenticate(req, res) if @auth
 
           res.status = 200
@@ -76,13 +91,14 @@ module PrometheusExporter::Server
           else
             res.body = metrics
           end
-        elsif req.path == '/send-metrics'
+        elsif req.path == "/send-metrics"
           handle_metrics(req, res)
-        elsif req.path == '/ping'
-          res.body = 'PONG'
+        elsif req.path == "/ping"
+          res.body = "PONG"
         else
           res.status = 404
-          res.body = "Not Found! The Prometheus Ruby Exporter only listens on /ping, /metrics and /send-metrics"
+          res.body =
+            "Not Found! The Prometheus Ruby Exporter only listens on /ping, /metrics and /send-metrics"
         end
       end
     end
@@ -94,13 +110,11 @@ module PrometheusExporter::Server
           @metrics_total.observe
           @collector.process(block)
         rescue => e
-          if @verbose
-            @logger.error "\n\n#{e.inspect}\n#{e.backtrace}\n\n"
-          end
+          @logger.error "\n\n#{e.inspect}\n#{e.backtrace}\n\n" if @verbose
           @bad_metrics_total.observe
           res.body = "Bad Metrics #{e}"
           res.status = e.respond_to?(:status_code) ? e.status_code : 500
-          return
+          break
         end
       end
 
@@ -109,13 +123,14 @@ module PrometheusExporter::Server
     end
 
     def start
-      @runner ||= Thread.start do
-        begin
-          @server.start
-        rescue => e
-          @logger.error "Failed to start prometheus collector web on port #{@port}: #{e}"
+      @runner ||=
+        Thread.start do
+          begin
+            @server.start
+          rescue => e
+            @logger.error "Failed to start prometheus collector web on port #{@port}: #{e}"
+          end
         end
-      end
     end
 
     def stop
@@ -125,7 +140,7 @@ module PrometheusExporter::Server
     def metrics
       metric_text = nil
       begin
-        Timeout::timeout(@timeout) do
+        Timeout.timeout(@timeout) do
           metric_text = @collector.prometheus_metrics_text
         end
       rescue Timeout::Error
@@ -158,9 +173,18 @@ module PrometheusExporter::Server
     end
 
     def get_rss
-      @pagesize ||= `getconf PAGESIZE`.to_i rescue 4096
+      @pagesize ||=
+        begin
+          `getconf PAGESIZE`.to_i
+        rescue StandardError
+          4096
+        end
       @pid ||= Process.pid
-      File.read("/proc/#{@pid}/statm").split(' ')[1].to_i * @pagesize rescue 0
+      begin
+        File.read("/proc/#{@pid}/statm").split(" ")[1].to_i * @pagesize
+      rescue StandardError
+        0
+      end
     end
 
     def add_gauge(name, help, value)
@@ -171,10 +195,12 @@ module PrometheusExporter::Server
 
     def authenticate(req, res)
       htpasswd = WEBrick::HTTPAuth::Htpasswd.new(@auth)
-      basic_auth = WEBrick::HTTPAuth::BasicAuth.new({ Realm: @realm, UserDB: htpasswd, Logger: @logger })
+      basic_auth =
+        WEBrick::HTTPAuth::BasicAuth.new(
+          { Realm: @realm, UserDB: htpasswd, Logger: @logger }
+        )
 
       basic_auth.authenticate(req, res)
     end
-
   end
 end

data/lib/prometheus_exporter/server.rb

@@ -19,3 +19,4 @@ require_relative "server/unicorn_collector"
 require_relative "server/active_record_collector"
 require_relative "server/shoryuken_collector"
 require_relative "server/resque_collector"
+require_relative "server/good_job_collector"

data/lib/prometheus_exporter/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module PrometheusExporter
-  VERSION = "2.0.8"
+  VERSION = "2.1.0"
 end

data/lib/prometheus_exporter.rb

@@ -2,21 +2,21 @@
 
 require_relative "prometheus_exporter/version"
 require "json"
-require "thread"
 
 module PrometheusExporter
   # per: https://github.com/prometheus/prometheus/wiki/Default-port-allocations
   DEFAULT_PORT = 9394
-  DEFAULT_BIND_ADDRESS = 'localhost'
-  DEFAULT_PREFIX = 'ruby_'
+  DEFAULT_BIND_ADDRESS = "localhost"
+  DEFAULT_PREFIX = "ruby_"
   DEFAULT_LABEL = {}
   DEFAULT_TIMEOUT = 2
-  DEFAULT_REALM = 'Prometheus Exporter'
+  DEFAULT_REALM = "Prometheus Exporter"
 
   class OjCompat
     def self.parse(obj)
       Oj.compat_load(obj)
     end
+
     def self.dump(obj)
       Oj.dump(obj, mode: :compat)
     end
@@ -25,7 +25,7 @@ module PrometheusExporter
   def self.hostname
     @hostname ||=
       begin
-        require 'socket'
+        require "socket"
         Socket.gethostname
       rescue => e
         STDERR.puts "Unable to lookup hostname #{e}"
@@ -45,13 +45,12 @@ module PrometheusExporter
   def self.has_oj?
     (
       @@has_oj ||=
-       begin
-         require 'oj'
-         :true
-       rescue LoadError
-         :false
-       end
-    ) == :true
+        begin
+          require "oj"
+          :T
+        rescue LoadError
+          :F
+        end
+    ) == :T
   end
-
 end

data/prometheus_exporter.gemspec

@@ -15,8 +15,6 @@ Gem::Specification.new do |spec|
   spec.homepage             = "https://github.com/discourse/prometheus_exporter"
   spec.license              = "MIT"
 
-  spec.post_install_message = "prometheus_exporter will only bind to localhost by default as of v0.5"
-
   spec.files = `git ls-files -z`.split("\x0").reject do |f|
     f.match(%r{^(test|spec|features|bin)/})
   end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: prometheus_exporter
 version: !ruby/object:Gem::Version
-  version: 2.0.8
+  version: 2.1.0
 platform: ruby
 authors:
 - Sam Saffron
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-01-20 00:00:00.000000000 Z
+date: 2024-01-07 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: webrick
@@ -243,7 +243,6 @@ extensions: []
 extra_rdoc_files: []
 files:
 - ".github/workflows/ci.yml"
-- ".github/workflows/docker.yml"
 - ".gitignore"
 - ".rubocop.yml"
 - Appraisals
@@ -261,11 +260,13 @@ files:
 - gemfiles/.bundle/config
 - gemfiles/ar_60.gemfile
 - gemfiles/ar_61.gemfile
+- gemfiles/ar_70.gemfile
 - lib/prometheus_exporter.rb
 - lib/prometheus_exporter/client.rb
 - lib/prometheus_exporter/instrumentation.rb
 - lib/prometheus_exporter/instrumentation/active_record.rb
 - lib/prometheus_exporter/instrumentation/delayed_job.rb
+- lib/prometheus_exporter/instrumentation/good_job.rb
 - lib/prometheus_exporter/instrumentation/hutch.rb
 - lib/prometheus_exporter/instrumentation/method_profiler.rb
 - lib/prometheus_exporter/instrumentation/periodic_stats.rb
@@ -290,6 +291,7 @@ files:
 - lib/prometheus_exporter/server/collector.rb
 - lib/prometheus_exporter/server/collector_base.rb
 - lib/prometheus_exporter/server/delayed_job_collector.rb
+- lib/prometheus_exporter/server/good_job_collector.rb
 - lib/prometheus_exporter/server/hutch_collector.rb
 - lib/prometheus_exporter/server/metrics_container.rb
 - lib/prometheus_exporter/server/process_collector.rb
@@ -311,8 +313,7 @@ homepage: https://github.com/discourse/prometheus_exporter
 licenses:
 - MIT
 metadata: {}
-post_install_message: prometheus_exporter will only bind to localhost by default as
-  of v0.5
+post_install_message: 
 rdoc_options: []
 require_paths:
 - lib

data/.github/workflows/docker.yml

@@ -1,47 +0,0 @@
-name: Docker
-
-on:
-  push:
-    tags:
-      - "v*"
-
-permissions:
-  contents: read
-  packages: write
-
-env:
-  DOCKER_REPO: ghcr.io/discourse/prometheus_exporter
-
-jobs:
-  build-and-publish:
-    runs-on: ubuntu-latest
-    timeout-minutes: 30
-
-    steps:
-      - name: Set vars
-        id: vars
-        run: |
-          echo "version=${GITHUB_REF#refs/tags/v}" >> $GITHUB_OUTPUT
-
-      - uses: actions/checkout@v3
-      - uses: docker/setup-qemu-action@v2
-      - uses: docker/setup-buildx-action@v2
-
-      - name: Login to Github Container Registry
-        uses: docker/login-action@v2
-        with:
-          registry: ghcr.io
-          username: ${{ github.actor }}
-          password: ${{ secrets.GITHUB_TOKEN }}
-
-      - name: Build and push images
-        uses: docker/build-push-action@v3
-        with:
-          context: .
-          push: true
-          platforms: linux/amd64,linux/arm64
-          build-args: |
-            GEM_VERSION=${{ steps.vars.outputs.version }}
-          tags: |
-            ${{ env.DOCKER_REPO }}:${{ steps.vars.outputs.version }}
-            ${{ env.DOCKER_REPO }}:latest