Arbiter.Worker.Config

Fields

• workerCount :: Int

  Number of concurrent worker threads.

• handlerMode :: HandlerMode m payload

  Job handler and claiming strategy. Set by transactionalWorkerConfig, manualWorkerConfig and batchedWorkerConfig.

• pollInterval :: NominalDiffTime

  Longest wait in seconds between dispatcher polls. A NOTIFY or a freed worker wakes the dispatcher sooner. Default: 5.

• visibilityTimeout :: NominalDiffTime

  The lease length: seconds a claim holds its job before another worker can claim it. Must be greater than jobHeartbeatInterval. Default: 60.

• jobHeartbeatInterval :: NominalDiffTime

  Interval in seconds for extending a job's lease during processing. Must be less than visibilityTimeout. Default: 30.

• maxJobDuration :: Maybe NominalDiffTime

  Interrupt a batch whose claim hooks and handler run longer than this many seconds. Its unfinalized jobs fail for retry. Nothing sets no bound. Default: Nothing.

• workerHeartbeatInterval :: NominalDiffTime

  Minimum gap in seconds between beats. A beat bumps the worker registry heartbeat, touches the optional liveness file, and reconciles pause state from the DB. Unless the pool is paused, a beat also waits for a dispatcher claim attempt or a lease extend. Keep it well below workerStaleThreshold. Default: 10.

• backoffStrategy :: BackoffStrategy

  Retry backoff strategy. Default: exponentialBackoff 2 1048576.

• jitter :: Jitter

  Jitter strategy for retry delays. Default: EqualJitter.

• observabilityHooks :: ObservabilityHooks m payload

  Callbacks for metrics or tracing. Default: defaultObservabilityHooks.

• onMaintenance :: MaintenanceOp -> Int64 -> m ()

  Called after a reaper op this pool won the gate for, with the rows it touched. Reaper work is schema-wide and carries no queue. Default: no-op.

• workerRuntime :: WorkerRuntime

  Mutable lifecycle state allocated for this pool.

• livenessFile :: Maybe FilePath

  When set, each beat touches this file, for a file-based liveness probe. See workerHeartbeatInterval for when a beat occurs. Default: arbiter-worker-<uuid> in the system temporary directory, from the UUID the constructor generated. Overriding workerId does not change it.

• gracefulShutdownTimeout :: Maybe NominalDiffTime

  Seconds a graceful shutdown waits on in-flight jobs before force-exiting. Nothing waits indefinitely. Default: Just 30.

• logConfig :: LogConfig

  Structured JSON log destination, level, and additional context. Default: Info to stdout, with a worker_id context pair from the UUID the constructor generated. Overriding workerId does not change it.

• cronJobs :: [CronJob payload]

  Cron schedules. A non-empty list gives the pool a scheduler thread, which reads the cron_schedules table each tick for runtime overrides. Default: [].

• reaperInterval :: NominalDiffTime

  Sleep in seconds between reaper passes, and the gap between runs of each ordinary reaper operation. Default: 300 (5 minutes).

• reaperSparseInterval :: NominalDiffTime

  Gap in seconds between runs of PruneRateLimitBuckets and ReconcilePruneConcurrency. Default: defaultMaintenanceSparseInterval.

• reaperBucketIdle :: NominalDiffTime

  Idle age in seconds at which the reaper prunes a full rate-limit bucket. Default: defaultMaintenanceBucketIdle.

• reaperTimeout :: NominalDiffTime

  Abort any single reaper statement that runs longer than this many seconds. Default: defaultMaintenanceTimeout.

• workerId :: UUID

  This worker's registry id. Default: a fresh random UUID.

• workerHost :: Maybe Text

  Hostname recorded in the worker registry. Default: this machine's hostname.

• workerMetadata :: Maybe Value

  Arbitrary JSONB metadata for the worker registry row (image tag, git SHA, deploy id, etc.). Default: Nothing.

• workerStaleThreshold :: NominalDiffTime

  The reaper's stale-worker sweep removes this worker's row when its heartbeat is older than this many seconds. Keep it well above workerHeartbeatInterval, pollInterval and jobHeartbeatInterval. Default: 300 (5 minutes).

Fields

• Int

  N, the maximum jobs in a batch.

• (NonEmpty (JobRead payload) -> BatchCallbacks m payload (ResultOf m payload) -> m ())
   

Fields

• ack :: JobRead payload -> m ()

  Ack and fire onJobSuccess, storing no result. Available on any queue. A job acked this way is absent from its parent rollup's child results and leaves its archive entry's result NULL.

• ackWith :: JobRead payload -> result -> m ()

  Ack, store the result for the parent rollup or the job's archive entry, fire onJobSuccess.

• ackAll :: [JobRead payload] -> m ()

  Bulk-ack in one parent-aware transaction, storing no results. Fires onJobSuccess per acked job. A job another worker holds is reported and skipped. The handler continues.

• ackAllWith :: [(JobRead payload, result)] -> m ()

  ackAll storing each job's result for its parent rollup or archive entry.

• failRetry :: JobRead payload -> Text -> m ()

  Retry with backoff, then DLQ at the job's maxAttempts. Fires onJobFailure, then onJobRetry or onJobFailedAndMovedToDLQ.

• failRetryAfter :: JobRead payload -> NominalDiffTime -> Text -> m ()

  failRetry after a delay in seconds in place of the backoff.

• failPermanent :: JobRead payload -> Text -> m ()

  Straight to the DLQ. Fires onJobFailure and onJobFailedAndMovedToDLQ.

• cancelBranch :: JobRead payload -> Text -> m ()

  Cancel the job's branch: its parent with all descendants, or the job's own subtree when it has no parent. Fires onJobCancelled.

• cancelTree :: JobRead payload -> Text -> m ()

  Cancel the whole tree from the root down. Fires onJobCancelled.

• nack :: JobRead payload -> m ()

  Reprocess after the lease expires. Records no failure and consumes no attempt. Fires no hook unless the job is no longer this worker's.

• nackAfter :: JobRead payload -> NominalDiffTime -> m ()

  nack that reprocesses after a delay in seconds in place of the lease's end.

• spawn :: JobRead payload -> NonEmpty (JobWrite payload) -> m ()

  Insert children under this job and suspend it, in one transaction. Fires onJobSuccess.