TaskForge
A production-grade, distributed job scheduler built to demonstrate resilient backend systems architecture and real-time dashboard tracking.
Try the "Chaos Demo"
TaskForge was built to be broken. Use the Observer dashboard to simulate crashes and see real-time resilience in action.
Spawn Jobs
Spawn 1 or 50 tasks to flood the system with jobs, flowing from Queued to Running to Completed.
Kill Worker
Inject a fatal crash into the Node.js worker process to simulate an ungraceful shutdown.
Observe Degraded State
Watch the dashboard Health Indicator instantly flip to Degraded as heartbeats halt.
Auto-Recovery
The Scheduler sweeps the DB, breaks stale locks, and re-queues jobs automatically.
Core Features
Distributed Execution
Horizontally scalable Node.js worker pool leveraging RabbitMQ for reliable message distribution.
Resilience & Chaos Engineering
Built-in 'Kill Worker' fault-injection to demonstrate automatic recovery from ungraceful shutdowns.
Real-Time Tracking
Next.js frontend using high-frequency SWR short-polling to render a responsive Kanban view.
Execution Guarantees
At-Least-Once execution guarantees with best-effort FIFO chronological ordering and backoff policies.
Global Event Console
Centralized polling logs capturing every event across the API, Scheduler, and Worker nodes.
Live System Health
Continuous system diagnostics on PostgreSQL, RabbitMQ, and Worker nodes with dynamic UI indicators.