rails_error_dashboard 0.10.0 → 0.11.1

data/config/locales/pt-BR.yml

@@ -147,6 +147,7 @@ pt-BR:
             any: Qualquer gravidade
           all_apps: Todas as aplicações
           all_platforms: Todas as plataformas
+          all_environments: Todos os ambientes
           all_types: Todos os tipos
           all_timeframes: Todo o período
           all_frequencies: Todas as frequências
@@ -175,6 +176,7 @@ pt-BR:
             search: 'Busca: %{value}'
             application: 'Aplicação: %{value}'
             platform: 'Plataforma: %{value}'
+            environment: 'Ambiente: %{value}'
             error_type: 'Tipo: %{value}'
             severity: 'Gravidade: %{value}'
             timeframe: 'Período: %{value}'
@@ -190,6 +192,7 @@ pt-BR:
           last_seen: Última ocorrência
           application: Aplicação
           platform: Plataforma
+          environment: Ambiente
           select_all: Selecionar todos os erros
         batch:
           selected:
@@ -650,6 +653,7 @@ pt-BR:
           low: BAIXA
         platform: Plataforma
         view_all_platform: Ver todos os erros de %{platform}
+        view_all_environment: Ver todos os erros de %{environment}
         user: Usuário
         view_user_errors: Ver todos os erros deste usuário
         user_id: 'ID: %{id}'
@@ -867,8 +871,9 @@ pt-BR:
         column_ips: IPs
         column_top_path: Caminho principal
         column_top_agent: Agente principal
-        overflow_notice_html: >-
-          Não foi possível atribuir %{count} eventos a nenhuma regra porque o buffer do rastreador estava cheio. Aumente %{option} para capturar mais.
+        overflow_notice_html: Não foi possível atribuir %{count} eventos a nenhuma
+          regra porque o buffer do rastreador estava cheio. Aumente %{option} para
+          capturar mais.
         match_type:
           blocklist: blocklist
           throttle: throttle
@@ -1107,6 +1112,9 @@ pt-BR:
         sampling_rate: Porcentagem de erros capturados
         application_name: Nome desta aplicação (detectado automaticamente se não for
           definido)
+        environment: Ambiente ao qual os erros são atribuídos (Rails.env se não definido)
+        notification_environments: Notificar apenas para estes ambientes (todos se
+          não definido)
         database: Banco de dados ativo para os logs de erros
         use_separate_database: Usar um banco de dados dedicado para os logs de erros
         user_model: Nome do modelo para as associações de usuário (detectado automaticamente
@@ -1277,11 +1285,14 @@ pt-BR:
         unknown_application: Desconhecida
       error_alert:
         subject: "\U0001F6A8 [%{application}] %{error_type}: %{message}"
+        subject_with_environment: "\U0001F6A8 [%{application} · %{environment}] %{error_type}:
+          %{message}"
         heading: "\U0001F6A8 Alerta de erro"
         labels:
           application: Aplicação
           error_type: Tipo de erro
           platform: Plataforma
+          environment: Ambiente
           occurred_at: Ocorreu em
           user_id: ID do usuário
           ip_address: Endereço IP
@@ -1350,6 +1361,7 @@ pt-BR:
           application: Aplicação
           error_type: Tipo de erro
           platform: Plataforma
+          environment: Ambiente
           occurred: Ocorreu
           message: Mensagem
           user: Usuário
@@ -1682,6 +1694,10 @@ pt-BR:
         platform_chart:
           title: Erros por plataforma
           quick_links: 'Links rápidos:'
+        environment_chart:
+          title: Erros por ambiente
+          quick_links: 'Links rápidos:'
+          unknown: Desconhecido
         type_chart:
           title: Os 10 tipos de erro mais frequentes
         hour_chart:

data/config/locales/ru.yml

@@ -147,6 +147,7 @@ ru:
             any: Любая важность
           all_apps: Все приложения
           all_platforms: Все платформы
+          all_environments: Все окружения
           all_types: Все типы
           all_timeframes: За всё время
           all_frequencies: Любая частота
@@ -175,6 +176,7 @@ ru:
             search: 'Поиск: %{value}'
             application: 'Приложение: %{value}'
             platform: 'Платформа: %{value}'
+            environment: 'Окружение: %{value}'
             error_type: 'Тип: %{value}'
             severity: 'Важность: %{value}'
             timeframe: 'Период: %{value}'
@@ -190,6 +192,7 @@ ru:
           last_seen: Последнее появление
           application: Приложение
           platform: Платформа
+          environment: Окружение
           select_all: Выбрать все ошибки
         batch:
           selected:
@@ -659,6 +662,7 @@ ru:
           low: НИЗКАЯ
         platform: Платформа
         view_all_platform: 'Все ошибки: %{platform}'
+        view_all_environment: 'Все ошибки: %{environment}'
         user: Пользователь
         view_user_errors: Все ошибки этого пользователя
         user_id: 'ID: %{id}'
@@ -879,8 +883,9 @@ ru:
         column_ips: IP-адреса
         column_top_path: Частый путь
         column_top_agent: Основной агент
-        overflow_notice_html: >-
-          Не удалось отнести %{count} событий ни к одному правилу, так как буфер трекера был заполнен. Увеличьте %{option}, чтобы записывать больше.
+        overflow_notice_html: Не удалось отнести %{count} событий ни к одному правилу,
+          так как буфер трекера был заполнен. Увеличьте %{option}, чтобы записывать
+          больше.
         match_type:
           blocklist: blocklist
           throttle: throttle
@@ -1136,6 +1141,9 @@ ru:
         sampling_rate: Доля захватываемых ошибок
         application_name: Имя этого приложения (определяется автоматически, если не
           задано)
+        environment: Окружение, к которому относятся ошибки (Rails.env, если не задано)
+        notification_environments: Уведомлять только для этих окружений (все, если
+          не задано)
         database: Активная база данных для журнала ошибок
         use_separate_database: Использовать отдельную БД для журнала ошибок
         user_model: Имя модели для связи с пользователями (определяется автоматически,
@@ -1309,11 +1317,14 @@ ru:
         unknown_application: Неизвестно
       error_alert:
         subject: "\U0001F6A8 [%{application}] %{error_type}: %{message}"
+        subject_with_environment: "\U0001F6A8 [%{application} · %{environment}] %{error_type}:
+          %{message}"
         heading: "\U0001F6A8 Оповещение об ошибке"
         labels:
           application: Приложение
           error_type: Тип ошибки
           platform: Платформа
+          environment: Окружение
           occurred_at: Время возникновения
           user_id: ID пользователя
           ip_address: IP-адрес
@@ -1385,6 +1396,7 @@ ru:
           application: Приложение
           error_type: Тип ошибки
           platform: Платформа
+          environment: Окружение
           occurred: Возникла
           message: Сообщение
           user: Пользователь
@@ -1715,6 +1727,10 @@ ru:
         platform_chart:
           title: Ошибки по платформам
           quick_links: 'Быстрые ссылки:'
+        environment_chart:
+          title: Ошибки по окружениям
+          quick_links: 'Быстрые ссылки:'
+          unknown: Неизвестно
         type_chart:
           title: Топ-10 типов ошибок
         hour_chart:

data/config/locales/uk.yml

@@ -147,6 +147,7 @@ uk:
             any: Будь-яка критичність
           all_apps: Усі застосунки
           all_platforms: Усі платформи
+          all_environments: Усі середовища
           all_types: Усі типи
           all_timeframes: За весь час
           all_frequencies: Усі частоти
@@ -175,6 +176,7 @@ uk:
             search: 'Пошук: %{value}'
             application: 'Застосунок: %{value}'
             platform: 'Платформа: %{value}'
+            environment: 'Середовище: %{value}'
             error_type: 'Тип: %{value}'
             severity: 'Критичність: %{value}'
             timeframe: 'Період: %{value}'
@@ -190,6 +192,7 @@ uk:
           last_seen: Востаннє помічено
           application: Застосунок
           platform: Платформа
+          environment: Середовище
           select_all: Вибрати всі помилки
         batch:
           selected:
@@ -657,6 +660,7 @@ uk:
           low: НИЗЬКА
         platform: Платформа
         view_all_platform: Переглянути всі помилки %{platform}
+        view_all_environment: Переглянути всі помилки %{environment}
         user: Користувач
         view_user_errors: Переглянути всі помилки цього користувача
         user_id: 'ID: %{id}'
@@ -877,8 +881,9 @@ uk:
         column_ips: IP-адреси
         column_top_path: Найчастіший шлях
         column_top_agent: Основний агент
-        overflow_notice_html: >-
-          Не вдалося віднести %{count} подій до жодного правила, оскільки буфер трекера був заповнений. Збільште %{option}, щоб записувати більше.
+        overflow_notice_html: Не вдалося віднести %{count} подій до жодного правила,
+          оскільки буфер трекера був заповнений. Збільште %{option}, щоб записувати
+          більше.
         match_type:
           blocklist: blocklist
           throttle: throttle
@@ -1132,6 +1137,9 @@ uk:
         sampling_rate: Відсоток захоплених помилок
         application_name: Назва цього застосунку (визначається автоматично, якщо не
           задано)
+        environment: Середовище, до якого належать помилки (Rails.env, якщо не задано)
+        notification_environments: Сповіщати лише для цих середовищ (усі, якщо не
+          задано)
         database: Активна база даних для журналів помилок
         use_separate_database: Використовувати окрему БД для журналів помилок
         user_model: Назва моделі для звʼязків із користувачами (визначається автоматично,
@@ -1307,11 +1315,14 @@ uk:
         unknown_application: Невідомо
       error_alert:
         subject: "\U0001F6A8 [%{application}] %{error_type}: %{message}"
+        subject_with_environment: "\U0001F6A8 [%{application} · %{environment}] %{error_type}:
+          %{message}"
         heading: "\U0001F6A8 Сповіщення про помилку"
         labels:
           application: Застосунок
           error_type: Тип помилки
           platform: Платформа
+          environment: Середовище
           occurred_at: Час виникнення
           user_id: ID користувача
           ip_address: IP-адреса
@@ -1383,6 +1394,7 @@ uk:
           application: Застосунок
           error_type: Тип помилки
           platform: Платформа
+          environment: Середовище
           occurred: Сталася
           message: Повідомлення
           user: Користувач
@@ -1714,6 +1726,10 @@ uk:
         platform_chart:
           title: Помилки за платформами
           quick_links: 'Швидкі посилання:'
+        environment_chart:
+          title: Помилки за середовищами
+          quick_links: 'Швидкі посилання:'
+          unknown: Невідомо
         type_chart:
           title: 10 найчастіших типів помилок
         hour_chart:

data/config/locales/zh-CN.yml

@@ -137,6 +137,7 @@ zh-CN:
             any: 任意严重程度
           all_apps: 全部应用
           all_platforms: 全部平台
+          all_environments: 全部环境
           all_types: 全部类型
           all_timeframes: 全部时间
           all_frequencies: 全部频率
@@ -165,6 +166,7 @@ zh-CN:
             search: 搜索：%{value}
             application: 应用：%{value}
             platform: 平台：%{value}
+            environment: 环境：%{value}
             error_type: 类型：%{value}
             severity: 严重程度：%{value}
             timeframe: 时间范围：%{value}
@@ -180,6 +182,7 @@ zh-CN:
           last_seen: 最近出现
           application: 应用
           platform: 平台
+          environment: 环境
           select_all: 选择全部错误
         batch:
           selected:
@@ -566,6 +569,7 @@ zh-CN:
           low: 低
         platform: 平台
         view_all_platform: 查看 %{platform} 的所有错误
+        view_all_environment: 查看 %{environment} 的所有错误
         user: 用户
         view_user_errors: 查看该用户的所有错误
         user_id: ID：%{id}
@@ -759,8 +763,7 @@ zh-CN:
         column_ips: IP 地址
         column_top_path: 主要路径
         column_top_agent: 主要代理
-        overflow_notice_html: >-
-          由于跟踪器缓冲区已满，%{count} 个事件无法归属到任何规则。请调高 %{option} 以捕获更多事件。
+        overflow_notice_html: 由于跟踪器缓冲区已满，%{count} 个事件无法归属到任何规则。请调高 %{option} 以捕获更多事件。
         match_type:
           blocklist: blocklist
           throttle: throttle
@@ -950,6 +953,8 @@ zh-CN:
         max_backtrace_lines: 堆栈跟踪深度上限
         sampling_rate: 错误捕获的采样比例
         application_name: 本应用的名称（未设置时自动检测）
+        environment: 错误所属的环境（未设置时为 Rails.env）
+        notification_environments: 仅对这些环境发送通知（未设置时为全部）
         database: 错误日志使用的数据库
         use_separate_database: 为错误日志使用独立数据库
         user_model: 用户关联所用的模型名（未设置时自动检测）
@@ -1087,11 +1092,13 @@ zh-CN:
         unknown_application: 未知
       error_alert:
         subject: "\U0001F6A8 [%{application}] %{error_type}：%{message}"
+        subject_with_environment: "\U0001F6A8 [%{application} · %{environment}] %{error_type}：%{message}"
         heading: "\U0001F6A8 错误告警"
         labels:
           application: 应用
           error_type: 错误类型
           platform: 平台
+          environment: 环境
           occurred_at: 发生时间
           user_id: 用户 ID
           ip_address: IP 地址
@@ -1154,6 +1161,7 @@ zh-CN:
           application: 应用
           error_type: 错误类型
           platform: 平台
+          environment: 环境
           occurred: 发生时间
           message: 消息
           user: 用户
@@ -1452,6 +1460,10 @@ zh-CN:
         platform_chart:
           title: 按平台统计错误
           quick_links: 快捷链接：
+        environment_chart:
+          title: 按环境统计的错误
+          quick_links: 快捷链接：
+          unknown: 未知
         type_chart:
           title: 出现最多的 10 种错误类型
         hour_chart:
@@ -1791,6 +1803,6 @@ zh-CN:
         axis_count: 数量
         axis_number_of_errors: 错误数量
         axis_hours: 小时
-        suffix_hours: "小时"
+        suffix_hours: 小时
         series_total_errors: 错误总数
         series_hours_to_resolve: 解决耗时（小时）

data/db/migrate/20260826000001_add_environment_to_error_logs.rb

@@ -0,0 +1,26 @@
+# frozen_string_literal: true
+
+# Records which environment an error came from (production, staging, uat, ...).
+#
+# RED had this column in v0.1.x and dropped it in the v1.0-launch refactor. It
+# returns as a nullable string: NULL means "captured before this column
+# existed", which is information the dashboard uses (no badge, matched as a
+# wildcard by FindOrIncrementError and adopted on the next occurrence), not a
+# gap to paper over with a default. `rails_error_dashboard:backfill_environments`
+# fills NULLs from environment_info.rails_env for anyone who wants history too.
+class AddEnvironmentToErrorLogs < ActiveRecord::Migration[7.0]
+  def change
+    # Guard against the squashed schema migration having already added this
+    # column -- without it, every later migration is silently cancelled.
+    return if column_exists?(:rails_error_dashboard_error_logs, :environment)
+
+    # 64 chars: long enough for "preprod-eu-west-2-canary", short enough that
+    # the composite index below stays cheap on MySQL's utf8mb4 key limit.
+    add_column :rails_error_dashboard_error_logs, :environment, :string, limit: 64
+
+    # Mirrors index_error_logs_on_platform_and_occurred_at: the index filter is
+    # "this environment, newest first".
+    add_index :rails_error_dashboard_error_logs, [ :environment, :occurred_at ],
+              name: "index_error_logs_on_environment_and_occurred_at"
+  end
+end

data/lib/generators/rails_error_dashboard/install/templates/initializer.rb

@@ -11,6 +11,17 @@ RailsErrorDashboard.configure do |config|
   config.dashboard_username = ENV.fetch("ERROR_DASHBOARD_USER", "gandalf")
   config.dashboard_password = ENV.fetch("ERROR_DASHBOARD_PASSWORD", "youshallnotpass")
 
+  # Environment awareness (v0.11.0). Every error records the environment it
+  # came from, the index can filter by it, and staging can be kept out of
+  # your pager. Defaults to Rails.env; override when a deploy runs under
+  # RAILS_ENV=production but is really "staging", "uat", "preprod"... names
+  # are free-form, never an enum.
+  # config.environment = ENV.fetch("ERROR_DASHBOARD_ENVIRONMENT", "staging")
+  #
+  # Only notify (Slack, email, Discord, PagerDuty, webhooks, storm and
+  # baseline alerts) for these environments. nil = every environment.
+  # config.notification_environments = %w[production]
+
   # === Custom Authentication (optional) ===
   # Use your app's existing auth instead of HTTP Basic Auth.
   # The lambda runs in controller context (via instance_exec), giving access to

data/lib/rails_error_dashboard/commands/backfill_environments.rb

@@ -0,0 +1,58 @@
+# frozen_string_literal: true
+
+module RailsErrorDashboard
+  module Commands
+    # Command: fill in `environment` on error logs captured before the column
+    # existed, using the rails_env recorded in environment_info at capture time.
+    #
+    # Opt-in (rails_error_dashboard:backfill_environments). Live errors do not
+    # need it — FindOrIncrementError adopts a NULL row on its next occurrence —
+    # but history that never recurs would otherwise stay unbadged and
+    # unfilterable forever.
+    #
+    # Runs in batches, one UPDATE per row: the value is inside a JSON text
+    # column, so it has to be parsed in Ruby and no portable single UPDATE can
+    # do it across PostgreSQL, MySQL and SQLite.
+    class BackfillEnvironments
+      BATCH_SIZE = 500
+      COLUMN_LIMIT = 64
+
+      def self.call(batch_size: BATCH_SIZE)
+        new(batch_size: batch_size).call
+      end
+
+      def initialize(batch_size: BATCH_SIZE)
+        @batch_size = batch_size
+      end
+
+      # @return [Integer] rows updated
+      def call
+        return 0 unless ErrorLog.column_names.include?("environment")
+
+        updated = 0
+        ErrorLog.where(environment: nil).where.not(environment_info: nil)
+                .in_batches(of: @batch_size) do |batch|
+          batch.pluck(:id, :environment_info).each do |id, raw|
+            env = rails_env_from(raw)
+            next if env.nil?
+
+            updated += ErrorLog.where(id: id, environment: nil).update_all(environment: env)
+          end
+        end
+        updated
+      end
+
+      private
+
+      def rails_env_from(raw)
+        parsed = JSON.parse(raw.to_s)
+        return nil unless parsed.is_a?(Hash)
+
+        value = parsed["rails_env"].to_s.strip
+        value.empty? ? nil : value[0, COLUMN_LIMIT]
+      rescue JSON::ParserError
+        nil
+      end
+    end
+  end
+end

data/lib/rails_error_dashboard/commands/find_or_increment_error.rb

@@ -9,7 +9,24 @@ module RailsErrorDashboard
     # 1. Unresolved errors with same hash within 24 hours → increment occurrence count
     # 2. Resolved/wont_fix errors with same hash (any age) → reopen and increment
     # 3. No match → create new error record
+    #
+    # Environment is a MATCH dimension, not part of the hash: the same error in
+    # staging and production is two rows with independent status. A row with a
+    # NULL environment predates the column; it matches as a wildcard and is
+    # stamped by the first occurrence that claims it, so history migrates
+    # itself without a backfill. An exact match always wins over a NULL one.
     class FindOrIncrementError
+      # Context that describes THIS occurrence rather than the error as a
+      # group. It is refreshed on every recurrence so the row always shows the
+      # latest moment of failure, not the first one in the 24 h window. Keys
+      # absent from @attributes (feature disabled, column not migrated, or a
+      # storm :lite capture that shed context) leave the stored payload alone —
+      # a shed capture must never blank out a good snapshot.
+      REFRESHED_CONTEXT = %i[
+        breadcrumbs system_health local_variables instance_variables
+        http_method hostname content_type request_duration_ms
+      ].freeze
+
       def self.call(error_hash, attributes = {})
         new(error_hash, attributes).call
       end
@@ -35,23 +52,47 @@ module RailsErrorDashboard
       private
 
       def find_unresolved
-        ErrorLog.unresolved
-          .where(error_hash: @error_hash)
-          .where(application_id: @attributes[:application_id])
-          .where("occurred_at >= ?", 24.hours.ago)
-          .lock
-          .order(last_seen_at: :desc)
-          .first
+        with_environment(
+          ErrorLog.unresolved
+            .where(error_hash: @error_hash)
+            .where(application_id: @attributes[:application_id])
+            .where("occurred_at >= ?", 24.hours.ago)
+        ).lock.order(last_seen_at: :desc).first
       end
 
       def find_resolved
-        ErrorLog
-          .where(error_hash: @error_hash)
-          .where(application_id: @attributes[:application_id])
-          .where(status: %w[resolved wont_fix])
-          .lock
-          .order(last_seen_at: :desc)
-          .first
+        with_environment(
+          ErrorLog
+            .where(error_hash: @error_hash)
+            .where(application_id: @attributes[:application_id])
+            .where(status: %w[resolved wont_fix])
+        ).lock.order(last_seen_at: :desc).first
+      end
+
+      # Restrict to this occurrence's environment or a legacy NULL row, exact
+      # first. Literal SQL, no interpolation. A blank environment (column not
+      # migrated yet, or an attribute-less caller) leaves the scope unchanged.
+      def with_environment(scope)
+        env = @attributes[:environment]
+        return scope if env.blank? || !ErrorLog.column_names.include?("environment")
+
+        scope.where(environment: [ env, nil ])
+             .order(Arel.sql("CASE WHEN environment IS NULL THEN 1 ELSE 0 END"))
+      end
+
+      # The subset of REFRESHED_CONTEXT this occurrence actually captured.
+      def latest_context
+        REFRESHED_CONTEXT.each_with_object({}) do |key, refreshed|
+          refreshed[key] = @attributes[key] unless @attributes[key].nil?
+        end
+      end
+
+      # {} unless this is a legacy NULL-environment row being claimed.
+      def environment_adoption(error)
+        return {} unless ErrorLog.column_names.include?("environment")
+        return {} if error.environment.present? || @attributes[:environment].blank?
+
+        { environment: @attributes[:environment] }
       end
 
       def increment_existing(error)
@@ -62,7 +103,9 @@ module RailsErrorDashboard
           request_url: @attributes[:request_url] || error.request_url,
           request_params: @attributes[:request_params] || error.request_params,
           user_agent: @attributes[:user_agent] || error.user_agent,
-          ip_address: @attributes[:ip_address] || error.ip_address
+          ip_address: @attributes[:ip_address] || error.ip_address,
+          **latest_context,
+          **environment_adoption(error)
         )
         error
       end
@@ -78,7 +121,9 @@ module RailsErrorDashboard
           request_url: @attributes[:request_url] || error.request_url,
           request_params: @attributes[:request_params] || error.request_params,
           user_agent: @attributes[:user_agent] || error.user_agent,
-          ip_address: @attributes[:ip_address] || error.ip_address
+          ip_address: @attributes[:ip_address] || error.ip_address,
+          **latest_context,
+          **environment_adoption(error)
         }
         attrs[:reopened_at] = Time.current if ErrorLog.column_names.include?("reopened_at")
         error.update!(attrs)
@@ -90,27 +135,29 @@ module RailsErrorDashboard
         ErrorLog.create!(@attributes.reverse_merge(resolved: false))
       rescue ActiveRecord::RecordNotUnique
         # Race condition: another process created the same error
-        retry_existing = ErrorLog.unresolved
-          .where(error_hash: @error_hash)
-          .where(application_id: @attributes[:application_id])
-          .where("occurred_at >= ?", 24.hours.ago)
-          .lock
-          .first
+        retry_existing = with_environment(
+          ErrorLog.unresolved
+            .where(error_hash: @error_hash)
+            .where(application_id: @attributes[:application_id])
+            .where("occurred_at >= ?", 24.hours.ago)
+        ).lock.first
 
         if retry_existing
           retry_existing.update!(
             occurrence_count: retry_existing.occurrence_count + 1,
-            last_seen_at: Time.current
+            last_seen_at: Time.current,
+            **latest_context,
+            **environment_adoption(retry_existing)
           )
           retry_existing
         else
           # Also check resolved in race condition path
-          retry_resolved = ErrorLog
-            .where(error_hash: @error_hash)
-            .where(application_id: @attributes[:application_id])
-            .where(status: %w[resolved wont_fix])
-            .lock
-            .first
+          retry_resolved = with_environment(
+            ErrorLog
+              .where(error_hash: @error_hash)
+              .where(application_id: @attributes[:application_id])
+              .where(status: %w[resolved wont_fix])
+          ).lock.first
 
           if retry_resolved
             reopen_existing(retry_resolved)