pg-boss-v12

Retry policies in a PostgreSQL job queue are configured using createQueue options, where you define retry strategies, set heartbeat intervals, and adjust expiration settings to control how failed background jobs are re-queued or dropped.

The difference between work and fetch lifecycles is that work() provides automatic job control by processing and completing jobs internally, while fetch() requires manual job handling where you explicitly call complete() to finish the background job.

You can schedule cron jobs and throttled tasks in Node.js using PostgreSQL by applying cron scheduling constraints and singleton, debounce, or throttle policies to control execution frequency and prevent overlapping background job runs.

Yes, pg-boss supports per-entity FIFO queues by applying stately queue policies, ensuring that background jobs for a specific entity are processed sequentially in the exact order they were sent to the PostgreSQL queue.

The best way to handle dead-letter queues and safe shutdown in PostgreSQL job queues is to configure dead-letter handling for exhausted retries and implement safe shutdown monitoring to ensure active jobs complete before the process terminates.

You chain background jobs for media transcoding pipelines in Node.js by using job chaining patterns to sequence tasks like probe, transcode, and thumbnail, dispatching follow-up jobs with boss.send() after each step completes.