500 Internal Server Error vs 502 Bad Gateway: Root Cause

Quick Decision

• -Use 500 when the responding server fails internally while processing the request.
• -Use 502 when a gateway/proxy receives an invalid response from upstream while acting as intermediary.
• -If the failure signature is at gateway-upstream boundary (parse/protocol/contract), classify as 502 first.

Key Differences

• -500 is an origin execution failure signal; 502 is an intermediary-upstream integration failure signal.
• -500 usually points to unhandled exceptions, internal dependency defects, or runtime instability in serving node.
• -502 usually points to malformed upstream responses, protocol mismatch, or gateway contract breakage.
• -Clear 500/502 separation reduces cross-team ping-pong during major incidents.

Step-by-Step Diagnosis

1. 1Trace one failing request end-to-end with shared correlation ID across edge, gateway, and origin.
2. 2Inspect gateway logs for upstream parse/protocol/TLS contract errors (candidate 502 path).
3. 3Inspect origin logs for exceptions, panic boundaries, and failed internal dependencies (candidate 500 path).
4. 4Compare direct-origin path versus gateway path to isolate whether the intermediary boundary introduces failure.

Common Misclassifications

• -Returning 500 for upstream-response invalidity and hiding gateway ownership.
• -Returning 502 for pure origin exception failures where no intermediary boundary is involved.
• -Treating all 5xx as one incident class and delaying handoff to the correct team.

Response Examples

HTTP/1.1 500 Internal Server Error
Content-Type: application/json

{
  "error": "internal_server_error",
  "request_id": "req_7f8a12",
  "message": "Unexpected exception while processing request."
}

HTTP/1.1 502 Bad Gateway
Content-Type: application/json

{
  "error": "bad_gateway",
  "upstream": "orders-service",
  "gateway": "edge-proxy-03",
  "message": "Gateway received an invalid upstream response."
}

Pro Tip

Frequently Asked Questions