A fast, reliable, and extensible distributed queue system built atop Redis, written in TypeScript.
Perfect for background job processing, such as email delivery, with built-in retries, scheduling, dead-letter queues, and metrics.
- Redis-powered queue: Reliable and fast task queue leveraging Redis lists and sorted sets
- Retry and dead-letter support: Automatic retries for failed tasks, with configurable dead-letter queue
- Task scheduling: Schedule tasks for future execution
- Worker-based processing: Scalable worker system to process tasks in parallel
- REST API producer: Simple HTTP API for enqueuing and monitoring tasks
- Metrics: Built-in queue and worker statistics
- TypeScript-first: Modern and type-safe
- Producer API: Accepts task enqueue requests via HTTP.
- Redis: Stores main queue, scheduled tasks, dead-letter queue, and metrics.
- Worker: Continuously processes and manages tasks, including retries and failures.
- Enqueue:
Producer API validates and pushes tasks to Redis (immediately or scheduled). - Worker Loop:
- Periodically moves due scheduled tasks to the main queue.
- Pops tasks from the main queue and processes them (e.g., sends email).
- On failure, retries up to max attempts, then moves to dead-letter queue.
- Updates metrics for processed/failed tasks.
- Monitoring:
Both API and worker expose metrics for queue health and throughput.
Create a .env file with your configuration:
REDIS_URL=
QUEUE_KEY=
DEAD_KEY=git clone https://github.com/Saloni1707/RedisDistro.git
cd RedisDistro
npm installnpm run producer
# Or: node src/producer/server.jsnpm run worker
# Or: node src/worker/worker.jsSend a POST request to enqueue an email job:
POST /enqueue/email
Content-Type: application/json
{
"to": "recipient@example.com",
"subject": "Your Subject",
"body": "Your message body",
"runAt": "tomorrow 8am"
}Response:
{ "message": "Task enqueued", "taskId": "..." }MIT © Saloni1707