good_job 4.19.0 → 4.19.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: d91dd892cc8a4110a371e8539b48200a20bb34bdbb08c3f5adc19ae96d2a9374
-  data.tar.gz: bea93501defb6c13acb563afced0a7ac340ce58f0567ad12143a7e82fbfd282f
+  metadata.gz: 6ad85ff0a870b41065fedac7c97b85f06606227b8a508851d775fabf90d2cc44
+  data.tar.gz: b4060b5afd727017bc337833d5c3d152f8f08465077edc42cbba9f929da72452
 SHA512:
-  metadata.gz: 2e4b30ddb0ef27b8409325dc98b775a46cb8398d5422c6387917f7d22df81ef5b5206ea4c23c9f8c2d9fd4e56b8ed0c27a46d26a021296e5f7c656ecb9ada865
-  data.tar.gz: 316857dc028b6d98db33e26fd68e5246659f0e79a92ae6fd923726e84770f8c069ef80ee73efdf608d478d4ed107f85a4afd4a0a45deb8145ed1147e43a9bf8a
+  metadata.gz: a6f6a889caf1e0566fb98b18603fabcfde6f2499cffc96d9ddec3eb3fd15e1049d32d2e16a0c42284410d43bd6bb849b57f5e54ac3dea989a1e9eba929a42034
+  data.tar.gz: f671e1ee8a746c4f771339d4f5afdd2da089e16934129462890645b44ebd013f13568feb1e278df06fb84ce49a1020e43f0813c481da60e16008d7b7944c05cf

data/CHANGELOG.md

@@ -1,8 +1,43 @@
 # Changelog
 
+## [v4.19.2](https://github.com/bensheldon/good_job/tree/v4.19.2) (2026-07-20)
+
+[Full Changelog](https://github.com/bensheldon/good_job/compare/v4.19.1...v4.19.2)
+
+**Fixed bugs:**
+
+- Resolve connection pool at call time in the `adapter_class` shim [\#1798](https://github.com/bensheldon/good_job/pull/1798) ([DaisyOli](https://github.com/DaisyOli))
+- Add GoodJob::Batch.where\(id:\) for GlobalID::Locator.fetch [\#1792](https://github.com/bensheldon/good_job/pull/1792) ([bensheldon](https://github.com/bensheldon))
+- Preserve created\_at when a job is retried [\#1789](https://github.com/bensheldon/good_job/pull/1789) ([ollym](https://github.com/ollym))
+
+**Closed issues:**
+
+- Memoized `adapter_class` proxy captures the pre-fork connection pool, crashing the Notifier in forked Puma workers \(Rails \< 7.2 + `preload_app!`\) [\#1797](https://github.com/bensheldon/good_job/issues/1797)
+- created\_at is reset every time a job is retried [\#1788](https://github.com/bensheldon/good_job/issues/1788)
+
+**Merged pull requests:**
+
+- Fix JRuby CI failure installing `rbs` gem [\#1794](https://github.com/bensheldon/good_job/pull/1794) ([bensheldon](https://github.com/bensheldon))
+- Fix flaky scheduled\_at precision comparison in adapter\_spec [\#1793](https://github.com/bensheldon/good_job/pull/1793) ([bensheldon](https://github.com/bensheldon))
+- Remove inline\_execution\_respects\_schedule config [\#1791](https://github.com/bensheldon/good_job/pull/1791) ([bensheldon](https://github.com/bensheldon))
+- Bump actions/checkout from 6 to 7 [\#1787](https://github.com/bensheldon/good_job/pull/1787) ([dependabot[bot]](https://github.com/apps/dependabot))
+
+## [v4.19.1](https://github.com/bensheldon/good_job/tree/v4.19.1) (2026-06-25)
+
+[Full Changelog](https://github.com/bensheldon/good_job/compare/v4.19.0...v4.19.1)
+
+**Fixed bugs:**
+
+- Support Rails Edge version of Arel::Table [\#1783](https://github.com/bensheldon/good_job/pull/1783) ([luizkowalski](https://github.com/luizkowalski))
+- Emit ActiveSupport::Notifications for concurrency-aborted enqueues [\#1779](https://github.com/bensheldon/good_job/pull/1779) ([bdewater-thatch](https://github.com/bdewater-thatch))
+
+**Closed issues:**
+
+- The probe server doesn't handle SIGINT/SIGTERM [\#1632](https://github.com/bensheldon/good_job/issues/1632)
+
 ## [v4.19.0](https://github.com/bensheldon/good_job/tree/v4.19.0) (2026-05-27)
 
-[Full Changelog](https://github.com/bensheldon/good_job/compare/v4.18.2...v4.19.0)
+[Full Changelog](https://github.com/bensheldon/good_job/compare/v4.18.3...v4.19.0)
 
 **Implemented enhancements:**
 
@@ -18,6 +53,17 @@
 - Make execution state completely Fiber-safe via Rails isolated execution state [\#1765](https://github.com/bensheldon/good_job/pull/1765) ([ollym](https://github.com/ollym))
 - Handle nil updated\_at in stale? method [\#1764](https://github.com/bensheldon/good_job/pull/1764) ([gavinballard](https://github.com/gavinballard))
 
+**Merged pull requests:**
+
+- Prune CI test matrix to boundary Ruby versions per Rails version [\#1777](https://github.com/bensheldon/good_job/pull/1777) ([bensheldon](https://github.com/bensheldon))
+- Refactor Concurrency::Rule to use explicit ivars instead of options hash [\#1776](https://github.com/bensheldon/good_job/pull/1776) ([bensheldon](https://github.com/bensheldon))
+- Fix typo in README.md [\#1773](https://github.com/bensheldon/good_job/pull/1773) ([NobodysNightmare](https://github.com/NobodysNightmare))
+- Use explicit keyword arguments for concurrency controls [\#1770](https://github.com/bensheldon/good_job/pull/1770) ([bdewater-thatch](https://github.com/bdewater-thatch))
+
+## [v4.18.3](https://github.com/bensheldon/good_job/tree/v4.18.3) (2026-05-27)
+
+[Full Changelog](https://github.com/bensheldon/good_job/compare/v4.18.2...v4.18.3)
+
 **Closed issues:**
 
 - enable\_pauses ENV var check has singular/plural mismatch [\#1771](https://github.com/bensheldon/good_job/issues/1771)
@@ -27,13 +73,6 @@
 - Job runner process enters a loop on create\_listen\_task - stale check fails [\#1649](https://github.com/bensheldon/good_job/issues/1649)
 - Set a default queue\_select\_limit [\#1596](https://github.com/bensheldon/good_job/issues/1596)
 
-**Merged pull requests:**
-
-- Prune CI test matrix to boundary Ruby versions per Rails version [\#1777](https://github.com/bensheldon/good_job/pull/1777) ([bensheldon](https://github.com/bensheldon))
-- Refactor Concurrency::Rule to use explicit ivars instead of options hash [\#1776](https://github.com/bensheldon/good_job/pull/1776) ([bensheldon](https://github.com/bensheldon))
-- Fix typo in README.md [\#1773](https://github.com/bensheldon/good_job/pull/1773) ([NobodysNightmare](https://github.com/NobodysNightmare))
-- Use explicit keyword arguments for concurrency controls [\#1770](https://github.com/bensheldon/good_job/pull/1770) ([bdewater-thatch](https://github.com/bdewater-thatch))
-
 ## [v4.18.2](https://github.com/bensheldon/good_job/tree/v4.18.2) (2026-04-20)
 
 [Full Changelog](https://github.com/bensheldon/good_job/compare/v4.18.1...v4.18.2)
@@ -117,6 +156,10 @@
 
 [Full Changelog](https://github.com/bensheldon/good_job/compare/v4.16.0...v4.17.0)
 
+**Implemented enhancements:**
+
+- Introduce advisory lock key customization support methods [\#1722](https://github.com/bensheldon/good_job/pull/1722) ([amkisko](https://github.com/amkisko))
+
 **Merged pull requests:**
 
 - Convert UI JavaScript modules to Stimulus controllers [\#1743](https://github.com/bensheldon/good_job/pull/1743) ([bensheldon](https://github.com/bensheldon))
@@ -132,6 +175,11 @@
 - Allow filtering by label on dashboard [\#1739](https://github.com/bensheldon/good_job/pull/1739) ([bensheldon](https://github.com/bensheldon))
 - Allow multiple concurrency rules per job via labels [\#1700](https://github.com/bensheldon/good_job/pull/1700) ([bscofield](https://github.com/bscofield))
 
+**Fixed bugs:**
+
+- Fix advisory lock connection stickiness in block contexts [\#1736](https://github.com/bensheldon/good_job/pull/1736) ([bensheldon](https://github.com/bensheldon))
+- Add JRuby 10 to testing matrix [\#1559](https://github.com/bensheldon/good_job/pull/1559) ([bensheldon](https://github.com/bensheldon))
+
 **Closed issues:**
 
 - Job duration misreported if interrupted [\#1723](https://github.com/bensheldon/good_job/issues/1723)
@@ -141,6 +189,7 @@
 - Use annotated git tag in release script [\#1741](https://github.com/bensheldon/good_job/pull/1741) ([bensheldon](https://github.com/bensheldon))
 - Double single-thread scheduler integration test timeout on JRuby [\#1738](https://github.com/bensheldon/good_job/pull/1738) ([bensheldon](https://github.com/bensheldon))
 - Fix JRuby test flakes for scheduler timeout and interrupted execution duration [\#1737](https://github.com/bensheldon/good_job/pull/1737) ([bensheldon](https://github.com/bensheldon))
+- Count Advisory Locks and refactor advisory lock lifecycle [\#1735](https://github.com/bensheldon/good_job/pull/1735) ([bensheldon](https://github.com/bensheldon))
 - Show interrupted execution recovery duration in dashboard [\#1733](https://github.com/bensheldon/good_job/pull/1733) ([bensheldon](https://github.com/bensheldon))
 - chore: use `merge` to avoid mutate the query object [\#1717](https://github.com/bensheldon/good_job/pull/1717) ([luizkowalski](https://github.com/luizkowalski))
 
@@ -156,6 +205,7 @@
 
 **Merged pull requests:**
 
+- Fix JRuby in development lockfile, with test [\#1734](https://github.com/bensheldon/good_job/pull/1734) ([bensheldon](https://github.com/bensheldon))
 - Add herb to linter [\#1732](https://github.com/bensheldon/good_job/pull/1732) ([bensheldon](https://github.com/bensheldon))
 - Update development dependencies; apply Rubocop to\_h lints [\#1728](https://github.com/bensheldon/good_job/pull/1728) ([bensheldon](https://github.com/bensheldon))
 
@@ -217,6 +267,7 @@
 
 **Fixed bugs:**
 
+- Check for graceful shutdown inside job cleanup loops [\#1711](https://github.com/bensheldon/good_job/pull/1711) ([bdewater-thatch](https://github.com/bdewater-thatch))
 - Add title to Good Job Dashboard layout [\#1701](https://github.com/bensheldon/good_job/pull/1701) ([mockdeep](https://github.com/mockdeep))
 
 **Closed issues:**
@@ -312,6 +363,7 @@
 **Merged pull requests:**
 
 - Update sorbet/tapioca [\#1681](https://github.com/bensheldon/good_job/pull/1681) ([bensheldon](https://github.com/bensheldon))
+- Remove obsolete property from tests [\#1676](https://github.com/bensheldon/good_job/pull/1676) ([RDIL](https://github.com/RDIL))
 - Bump actions/checkout from 4 to 5 [\#1673](https://github.com/bensheldon/good_job/pull/1673) ([dependabot[bot]](https://github.com/apps/dependabot))
 
 ## [v4.11.2](https://github.com/bensheldon/good_job/tree/v4.11.2) (2025-08-06)
@@ -434,6 +486,10 @@
 
 [Full Changelog](https://github.com/bensheldon/good_job/compare/v4.9.0...v4.9.1)
 
+**Implemented enhancements:**
+
+- Order Dashboard jobs in more "natural" order [\#1604](https://github.com/bensheldon/good_job/pull/1604) ([francois](https://github.com/francois))
+
 **Fixed bugs:**
 
 - \[dashboard\] Scheduled tasks are shown "backwards" [\#1580](https://github.com/bensheldon/good_job/issues/1580)
@@ -1919,7 +1975,6 @@
 - In tests, directly use `connection.truncate_tables` instead of  `ActiveRecord::Tasks::DatabaseTasks.truncate_all` [\#918](https://github.com/bensheldon/good_job/pull/918) ([bensheldon](https://github.com/bensheldon))
 - Use `GoodJob.capsule` in CLI instead of creating a new Capsule [\#916](https://github.com/bensheldon/good_job/pull/916) ([bensheldon](https://github.com/bensheldon))
 - Fix lint: missing fr and nl translations; readme whitespace [\#914](https://github.com/bensheldon/good_job/pull/914) ([bensheldon](https://github.com/bensheldon))
-- Switches from Chrome to Chromium to support Apple M1/Arm CPUs [\#910](https://github.com/bensheldon/good_job/pull/910) ([ckdake](https://github.com/ckdake))
 - Bump erb\_lint from 0.3.1 to 0.4.0 [\#906](https://github.com/bensheldon/good_job/pull/906) ([dependabot[bot]](https://github.com/apps/dependabot))
 
 ## [v3.15.0](https://github.com/bensheldon/good_job/tree/v3.15.0) (2023-04-02)

data/README.md

@@ -315,7 +315,6 @@ Available configuration options are:
 - `cleanup_preserved_jobs_before_seconds_ago` (integer) number of seconds to preserve jobs when using the `$ good_job cleanup_preserved_jobs` CLI command or calling `GoodJob.cleanup_preserved_jobs`. Defaults to `1209600` (14 days). Can also be set with  the environment variable `GOOD_JOB_CLEANUP_PRESERVED_JOBS_BEFORE_SECONDS_AGO`.
 - `cleanup_interval_jobs` (integer) Number of jobs a Scheduler will execute before cleaning up preserved jobs. Defaults to `1000`. Disable with `false`. Can also be set with  the environment variable `GOOD_JOB_CLEANUP_INTERVAL_JOBS` and disabled with `0`).
 - `cleanup_interval_seconds` (integer) Number of seconds a Scheduler will wait before cleaning up preserved jobs. Defaults to `600` (10 minutes). Disable with `false`. Can also be set with  the environment variable `GOOD_JOB_CLEANUP_INTERVAL_SECONDS` and disabled with `0`).
-- `inline_execution_respects_schedule` (boolean) Opt-in to future behavior of inline execution respecting scheduled jobs. Defaults to `false`.
 - `logger` ([Rails Logger](https://api.rubyonrails.org/classes/ActiveSupport/Logger.html)) lets you set a custom logger for GoodJob. It should be an instance of a Rails `Logger` (Default: `Rails.logger`).
 - `preserve_job_records` (boolean, symbol, or lambda) keeps job records in your database even after jobs are completed. If set to `true`, all job records are preserved. If set to `:on_unhandled_error`, only jobs that finished with an unhandled error are preserved. If set to a lambda, the lambda will be called with the error_event (e.g., `:discarded`, `:retry_stopped`, or `:unhandled`) and should return a boolean indicating whether to preserve the job. (Default: `true`)
 - `advisory_lock_heartbeat` (boolean) whether to use an advisory lock for the purpose of determining whether an execeution process is active. (Default `true` in Development; `false` in other environments)
@@ -339,6 +338,9 @@ Available configuration options are:
     ```
 
 - `enable_pauses` (boolean) whether job processing can be paused. Defaults to `false`. You can also set this with the environment variable `GOOD_JOB_ENABLE_PAUSES`.
+- `dequeue_query_sort` (symbol) sets the sort order used when querying for the next job(s) to execute. Defaults to `:created_at`. It can be any one of:
+    - `:created_at` sorts by priority, then job creation order. This is the default, but no longer best matches GoodJob's contemporary data model.
+    - `:scheduled_at` sorts by priority, then `scheduled_at`, then `id`, which better matches GoodJob's contemporary data model. This will become the default/only behavior in a future major release.
 
 By default, GoodJob configures the following execution modes per environment:
 
@@ -1505,9 +1507,7 @@ It is also possible to manually trigger a cleanup of preserved job records:
 
 By default, GoodJob uses its inline adapter in the test environment; the inline adapter is designed for the test environment. When enqueuing a job with GoodJob's inline adapter, the job will be executed immediately on the current thread; unhandled exceptions will be raised.
 
-In GoodJob 2.0, the inline adapter will execute future scheduled jobs immediately. In the next major release, GoodJob 3.0, the inline adapter will not execute future scheduled jobs and instead enqueue them in the database.
-
-To opt into this behavior immediately set: `config.good_job.inline_execution_respects_schedule = true`
+The inline adapter respects scheduled jobs: jobs scheduled for the future are enqueued in the database rather than executed immediately.
 
 To perform jobs inline at any time, use `GoodJob.perform_inline`. For example, using time helpers within an integration test:
 

data/app/models/concerns/good_job/advisory_lockable.rb

@@ -207,6 +207,7 @@ module GoodJob
     end
 
     ADVISORY_LOCK_COUNTS = AdvisoryLockCounter.new
+    AREL_TABLE_NEW_KWARGS = Arel::Table.instance_method(:initialize).parameters.any? { |type, name| type == :key && name == :name }
 
     included do
       # Default column to be used when creating Advisory Locks
@@ -224,10 +225,10 @@ module GoodJob
       unless respond_to?(:adapter_class)
         define_singleton_method(:adapter_class) do
           @_adapter_class ||= begin
-            pool = connection_pool
+            klass = self
             proxy = Object.new
-            proxy.define_singleton_method(:quote_table_name) { |name| pool.with_connection { |c| c.quote_table_name(name) } }
-            proxy.define_singleton_method(:quote_column_name) { |name| pool.with_connection { |c| c.quote_column_name(name) } }
+            proxy.define_singleton_method(:quote_table_name) { |name| klass.connection_pool.with_connection { |c| c.quote_table_name(name) } }
+            proxy.define_singleton_method(:quote_column_name) { |name| klass.connection_pool.with_connection { |c| c.quote_column_name(name) } }
             proxy
           end
         end
@@ -357,7 +358,7 @@ module GoodJob
           primary_key_for_select = primary_key.to_sym
           column_for_select = column.to_sym
 
-          cte_table = Arel::Table.new(:rows)
+          cte_table = arel_table_for(:rows)
           cte_query = original_query.except(:limit)
           cte_query = if primary_key_for_select == column_for_select
                         cte_query.select(primary_key_for_select)
@@ -485,6 +486,13 @@ module GoodJob
 
       private
 
+      # Arel::Table.new has different parameters depending on Rails version.
+      # New version expects keyword arguments for the name while the old version expects a positional argument.
+      # This method creates a new Arel::Table instance with the correct parameters.
+      def arel_table_for(name)
+        AREL_TABLE_NEW_KWARGS ? Arel::Table.new(name: name) : Arel::Table.new(name)
+      end
+
       # Executes a single pg_advisory_unlock call and updates bookkeeping.
       # Used by {.advisory_unlock_key} and {.advisory_unlock_key!} to avoid
       # mutual recursion when CTE-aware cleanup is needed.

data/app/models/good_job/batch.rb

@@ -197,6 +197,12 @@ module GoodJob
       new _record: BatchRecord.find(id)
     end
 
+    # Required by GlobalID::Locator.fetch/locate_many, which query with
+    # `ignore_missing: true` and thus call `.where(id: ids)` instead of `.find(id)`.
+    def self.where(id:)
+      BatchRecord.where(id: id).map { |record| new(_record: record) }
+    end
+
     # Helper method to enqueue jobs and assign them to a batch
     def self.within_thread(batch_id: nil, batch_callback_id: nil)
       original_batch_id = current_batch_id

data/app/models/good_job/job.rb

@@ -327,8 +327,11 @@ module GoodJob
       new(**enqueue_args(active_job, scheduled_at: scheduled_at))
     end
 
-    # Construct arguments for GoodJob::Execution from an ActiveJob instance.
+    # Construct arguments for GoodJob::Job from an ActiveJob instance.
     def self.enqueue_args(active_job, scheduled_at: nil)
+      reenqueued_current_job = CurrentThread.active_job_id && CurrentThread.active_job_id == active_job.job_id
+      current_job = CurrentThread.job
+
       execution_args = {
         id: active_job.job_id,
         active_job_id: active_job.job_id,
@@ -336,15 +339,16 @@ module GoodJob
         queue_name: active_job.queue_name.presence || DEFAULT_QUEUE_NAME,
         priority: active_job.priority || DEFAULT_PRIORITY,
         serialized_params: active_job.serialize,
-        created_at: Time.current,
       }
 
+      now = Time.current
+      execution_args[:created_at] = now if !reenqueued_current_job || GoodJob.configuration.dequeue_query_sort == :created_at
       execution_args[:scheduled_at] = if scheduled_at
                                         scheduled_at
                                       elsif active_job.scheduled_at
                                         Time.zone.at(active_job.scheduled_at)
                                       else
-                                        execution_args[:created_at]
+                                        now
                                       end
 
       execution_args[:concurrency_key] = active_job.good_job_concurrency_key if active_job.respond_to?(:good_job_concurrency_key)
@@ -356,9 +360,6 @@ module GoodJob
         execution_args[:labels] = labels
       end
 
-      reenqueued_current_job = CurrentThread.active_job_id && CurrentThread.active_job_id == active_job.job_id
-      current_job = CurrentThread.job
-
       if reenqueued_current_job
         execution_args[:batch_id] = current_job.batch_id
         execution_args[:batch_callback_id] = current_job.batch_callback_id

data/config/brakeman.ignore

@@ -66,13 +66,13 @@
     {
       "warning_type": "Dangerous Eval",
       "warning_code": 13,
-      "fingerprint": "c4c3e1b8b28ebbfd4672cf3e8f0022a27aff3c12dc4fea750b412de1387f91e6",
+      "fingerprint": "c01ff32ee40cc694540aaa76e3fdab8f330babd9d6dbf2564c32aef1cfd877fa",
       "check_name": "Evaluation",
       "message": "Dynamic string evaluated as code",
       "file": "lib/good_job/log_subscriber.rb",
-      "line": 256,
+      "line": 278,
       "link": "https://brakemanscanner.org/docs/warning_types/dangerous_eval/",
-      "code": "class_eval(\"        def #{level}(progname = nil, tags: [], &block)   # def info(progname = nil, tags: [], &block)\\n          return unless logger                           #   return unless logger\\n                                                         #\\n          tag_logger(*tags) do                           #   tag_logger(*tags) do\\n            logger.#{level}(progname, &block)            #     logger.info(progname, &block)\\n          end                                            #   end\\n        end                                              #\\n\", \"lib/good_job/log_subscriber.rb\", (256 + 1))",
+      "code": "class_eval(\"        def #{level}(progname = nil, tags: [], &block)   # def info(progname = nil, tags: [], &block)\\n          return unless logger                           #   return unless logger\\n                                                         #\\n          tag_logger(*tags) do                           #   tag_logger(*tags) do\\n            logger.#{level}(progname, &block)            #     logger.info(progname, &block)\\n          end                                            #   end\\n        end                                              #\\n\", \"lib/good_job/log_subscriber.rb\", (278 + 1))",
       "render_path": null,
       "location": {
         "type": "method",

data/lib/good_job/active_job_extensions/concurrency.rb

@@ -120,7 +120,10 @@ module GoodJob
                                       end
 
                 if (enqueue_concurrency + 1) > limit
-                  job.logger.info "Aborted enqueue of #{job.class.name} (Job ID: #{job.job_id}) because the concurrency key '#{key}' has reached its enqueue limit of #{limit} #{'job'.pluralize(limit)}"
+                  ActiveSupport::Notifications.instrument(
+                    "enqueue_concurrency_limit_exceeded.good_job",
+                    { job: job, key: key, limit: limit }
+                  )
                   exceeded = :limit
                   next
                 end
@@ -134,7 +137,10 @@ module GoodJob
                                          .count
 
                 if (enqueued_within_period + 1) > throttle_limit
-                  job.logger.info "Aborted enqueue of #{job.class.name} (Job ID: #{job.job_id}) because the concurrency key '#{key}' has reached its throttle limit of #{throttle_limit} #{'job'.pluralize(throttle_limit)}"
+                  ActiveSupport::Notifications.instrument(
+                    "enqueue_concurrency_throttle_exceeded.good_job",
+                    { job: job, key: key, limit: throttle_limit }
+                  )
                   exceeded = :throttle
                   next
                 end

data/lib/good_job/configuration.rb

@@ -177,10 +177,6 @@ module GoodJob
       end
     end
 
-    def inline_execution_respects_schedule?
-      !!rails_config[:inline_execution_respects_schedule]
-    end
-
     # The maximum number of future-scheduled jobs to store in memory.
     # Storing future-scheduled jobs in memory reduces execution latency
     # at the cost of increased memory usage. 10,000 stored jobs = ~20MB.
@@ -439,6 +435,7 @@ module GoodJob
       Rails.env.development?
     end
 
+    # TODO: Remove :created_at and this option entirely in the next major version; :scheduled_at will become the only behavior.
     def dequeue_query_sort
       (options[:dequeue_query_sort] || rails_config[:dequeue_query_sort] || :created_at).to_sym
     end

data/lib/good_job/log_subscriber.rb

@@ -153,6 +153,28 @@ module GoodJob
       end
     end
 
+    # @!macro notification_responder
+    def enqueue_concurrency_limit_exceeded(event)
+      job = event.payload[:job]
+      key = event.payload[:key]
+      limit = event.payload[:limit]
+
+      info do
+        "Aborted enqueue of #{job.class.name} (Job ID: #{job.job_id}) because the concurrency key '#{key}' has reached its enqueue limit of #{limit} #{'job'.pluralize(limit)}"
+      end
+    end
+
+    # @!macro notification_responder
+    def enqueue_concurrency_throttle_exceeded(event)
+      job = event.payload[:job]
+      key = event.payload[:key]
+      limit = event.payload[:limit]
+
+      info do
+        "Aborted enqueue of #{job.class.name} (Job ID: #{job.job_id}) because the concurrency key '#{key}' has reached its throttle limit of #{limit} #{'job'.pluralize(limit)}"
+      end
+    end
+
     # @!macro notification_responder
     def systemd_watchdog_start(event)
       interval = event.payload[:interval]

data/lib/good_job/version.rb

@@ -2,7 +2,7 @@
 
 module GoodJob
   # GoodJob gem version.
-  VERSION = '4.19.0'
+  VERSION = '4.19.2'
 
   # GoodJob version as Gem::Version object
   GEM_VERSION = Gem::Version.new(VERSION)

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: good_job
 version: !ruby/object:Gem::Version
-  version: 4.19.0
+  version: 4.19.2
 platform: ruby
 authors:
 - Ben Sheldon