**Project:** focused-intuition (0814070e-b62e-4ac3-8366-7678e492f30b)

**Service:** @plataforma/api (348e032b-d53e-4a95-8292-892afd810e4a)

**Environment:** dev (0265e4a3-da7d-47e4-8ed1-79400b9c2997)

**Region:** us-east4-eqdc4a

**Problem:**

The container starts ("Starting Container" log entry) but never outputs any Node.js/NestJS logs and returns 502 for all HTTP requests. This started around 11:00 AM CST March 15, 2026 after several redeployments triggered by env var changes.

**Key facts:**

- The exact same Docker image works perfectly in our production environment (same project, different Railway environment - production)

- Memory usage is stable at ~100MB per Metrics panel (not OOM)

- Deploy logs only show "Starting Container" - no Node.js output at all

- Rollbacks to previously working commits also fail the same way

- The deployment status shows as "Deployment successful" / ACTIVE but the container returns 502

- We can see SIGTERM signals being sent, suggesting Railway is cycling the container

**What we've tried:**

- Multiple rollbacks to known-good commits

- Changing NODE_ENV env var (triggered redeployments)

- Adding healthcheck path /api/health

- The issue persists across all deployment attempts since ~11 AM today

Could this be a platform-level issue in the us-east4-eqdc4a region or with this specific service instance? The production environment in the same region works fine.