503 Service Unavailable Checklist

Problem

The final response code is 503 Service Unavailable, indicating the service cannot currently handle the request.

Symptoms

• Browsers show a Service Unavailable or maintenance-style page.
• HTTP Check shows a final status code of 503.
• The issue is more visible during traffic spikes or on specific routes.

Top 3 Causes

1. The service is overloaded - The application or infrastructure lacks available capacity.
2. Maintenance or protection mode is active - Deploy locks, maintenance pages, WAF protections, or rate limits are rejecting traffic.
3. Autoscaling or dependency readiness is delayed - Backends are not ready quickly enough to serve requests.

Diagnose with DechoNet

• HTTP Check to inspect the final status code, headers, and any maintenance pattern.
• Port Check to confirm the primary service port still responds externally.

Resolution Checklist

• Check whether maintenance mode or deployment locks are active.
• Review rate limiting, bot protection, or overload protection policies that can emit 503.
• Inspect application capacity, autoscaling behavior, and resource usage.
• Verify that dependencies such as databases, caches, or upstream APIs are healthy.
• Re-run HTTP Check after changes and confirm the final code no longer returns 503.

When to Escalate

• Escalate if the issue comes from shared infrastructure or platform-wide overload.
• Repeated 503 incidents usually require capacity planning and better operational alerting, not just a one-time fix.