For LiveKit workers running on ECS, if your health checks are failing due to slow worker startup, it is recommended to extend the ECS health check start period, timeout, and retries to allow more time for the workers to fully initialize. Here are the recommended ECS health check settings based on LiveKit deployment best practices: • command:

["CMD-SHELL", "curl -f <http://localhost:8081/> || exit 1"]

• interval:

30

(seconds) • timeout:

10-15

(seconds) • retries:

5-6

• startPeriod:

60-120

(seconds, increase for slow-starting workers) Additionally, ensure your ECS task definition includes a suitable stopTimeout (graceful shutdown period). For standard workers, use

300

seconds, and for voice agents or long-running jobs, consider

600

seconds or more. These adjustments ensure workers have sufficient time to start before ECS considers them unhealthy and help agents complete tasks during shutdowns. Let me know if you need example ECS task definition configuration or further details for your setup. Sources: livekit-agents/livekit/agents/ipc/supervised_proc.py | WorkerOptions | LiveKit Agents | Deployment and scaling | LiveKit Docs | fly.io/fly.toml | Deploying to production | LiveKit Docs