Receiving a text message that a job failed is good. Not having to wake up because the system fixed itself is better.
Closing the Loop with Webhooks
CronRabbit can trigger webhooks on failure. Instead of just notifying a human, you can point that webhook at:
- GitHub Actions: Re-run a failed CI deploy.
- Kubernetes: Scale a deployment to 0 then back to 1 to clear a memory leak.
- AWS Lambda: Run a diagnostic script.
The "Health Check" Restart
If a worker stops heartbeating, there's a high chance it's "zombie" (alive but not doing work). Automated remediation ensures your system stays healthy while you sleep.