How to Troubleshoot HTTP 500, 502, and 503

5xx errors mean the server side failed, but each code points to a different failure layer. This guide helps you isolate the layer quickly.

Fast triage (first 5 minutes)

1. Run the URL in Website Checker to capture status and timing.
2. Confirm whether the issue affects one URL path or all endpoints.
3. Check if the error is stable or intermittent.
4. Correlate with deploys, config changes, or traffic spikes.

500 Internal Server Error

A generic app/server failure. Most common causes: uncaught application exception, bad environment variable, permission issue, or a failed dependency call.

• Review application logs first, then web server logs.
• Check recent releases and migrations.
• Verify file permissions and runtime secrets.

502 Bad Gateway

Usually a proxy/gateway issue between Nginx/Apache/load balancer and your upstream app.

• Confirm upstream process is running and listening on expected host/port.
• Check proxy timeout and keepalive mismatches.
• Validate upstream DNS in containerized deployments.

503 Service Unavailable

Service temporarily unavailable, often due to overload, maintenance mode, or exhausted worker pools.

• Inspect CPU, memory, and connection pool saturation.
• Check autoscaling events and rate limiting.
• Verify maintenance flags or misconfigured health checks.

What to send in support escalations

• URL + exact status code
• First observed timestamp (UTC preferred)
• Whether issue is global or region-specific
• Any recent deploy or infrastructure changes

For DNS-side failures, see our DNS propagation guide. For certificate host mismatch failures, see this SSL guide.