n8n-io/n8n · warning

Bad Gateway

Error message

Bad Gateway

What it means

The telemetry/source-config controller proxies requests to api-rs.n8n.io. The http-proxy 'error' callback writes a raw 502 response with body 'Bad Gateway' whenever the proxy itself fails to reach the upstream (before any HTTP response is received).

Source

Thrown at packages/cli/src/controllers/telemetry.controller.ts:44

					return;
				},
				proxyRes: (proxyRes) => {
					// MCP app UIs call this cross-origin from sandboxed iframes. Upstream
					// may set CORS headers too, so normalize to one permissive value.
					for (const header of [
						'access-control-allow-origin',
						'access-control-allow-credentials',
						'access-control-allow-methods',
						'access-control-allow-headers',
						'access-control-expose-headers',
					]) {
						delete proxyRes.headers[header];
					}
					proxyRes.headers['access-control-allow-origin'] = '*';
				},
				error: (_error, _req, res) => {
					if ('writeHead' in res && !res.headersSent) {
						res.writeHead(502, { 'Access-Control-Allow-Origin': '*' });
						res.end('Bad Gateway');
					}
				},
			},
		});
	}

	private applyCors(req: AuthenticatedRequest, res: Response) {
		res.setHeader('Access-Control-Allow-Origin', '*');
		res.setHeader('Access-Control-Allow-Methods', 'GET, POST, OPTIONS');

		const requestedHeaders = req.headers['access-control-request-headers'];
		res.setHeader(
			'Access-Control-Allow-Headers',
			typeof requestedHeaders === 'string' && requestedHeaders.length > 0
				? requestedHeaders
				: 'Content-Type, Authorization, anonymousId',
		);

View on GitHub (pinned to 5ac6606e81)

Solutions

1. Allow outbound HTTPS to api-rs.n8n.io (and the RudderStack data plane) in firewall/egress rules.
2. If the instance is offline, disable diagnostics (N8N_DIAGNOSTICS_ENABLED=false) so the proxy is not exercised.
3. Route n8n through a corporate forward proxy (HTTP_PROXY/HTTPS_PROXY) if direct egress is blocked.

Example fix

// before: offline/air-gapped instance still trying to proxy -> 502 Bad Gateway
// after: disable diagnostics for offline instances
// .env
N8N_DIAGNOSTICS_ENABLED=false

// Skip telemetry calls on offline instances.
function shouldProxyTelemetry(env = process.env) {
  return String(env.N8N_DIAGNOSTICS_ENABLED).toLowerCase() !== 'false';
}

// The 502 is a raw response, not a thrown JS error. Handle at the HTTP layer.
if (res.status === 502 && res.body === 'Bad Gateway') {
  // degrade telemetry silently; do not block the UI
}

When it happens

Trigger: Any proxied telemetry/source-config request where the proxy emits an error event — upstream unreachable, DNS failure, connection reset, TLS handshake failure — and res.headersSent is still false.

Common situations: Air-gapped/offline instance; firewall/egress rules block outbound to api-rs.n8n.io; DNS misconfiguration; corporate proxy not configured for n8n.

Understand the failure class

• HTTP status errors: handling 4xx and 5xx responses — how to handle 4xx and 5xx responses properly.

Related errors

• Failed to fetch source config: ${response.statusMessage}
• Failed to fetch provider catalog: ${response.statusText}
• OpenTelemetry active span callback is required.
• Telemetry tracer must implement startSpan() and startActiveS
• Cannot set both .tracer() and .otlpEndpoint() — use one or t

AI-assisted analysis of n8n-io/n8n@5ac6606e81 (2026-08-12). Data as JSON: /api/errors/01fe99c657be08ac. Report an issue: GitHub.