Keep dashboard control failures contained

Proof: Live dashboard preflight returned an empty client response because an upstream control timeout rejected outside the server catch path and restarted operator-dashboard. The server now awaits API handlers, wraps control proxy failures, and returns structured JSON errors instead of crashing.

Assumptions: A timed-out control proxy should be observable as a dashboard control failure, not as process death. Returning 504/502 JSON is safer than treating the action as successful or hiding the upstream state.

Still fake: This does not make unanswered solver quote requests fill; it only makes the dashboard control surface truthful and non-crashing for failed or timed-out controls.

src/apps/operator-dashboard.mjs

@@ -10,6 +10,7 @@ import { parseEventMessage } from '../core/event-envelope.mjs';
 import {
   applyDashboardLiveEvent,
   buildDashboardBootstrap,
+  buildDashboardControlErrorResponse,
   buildLiveStatusBar,
   createDashboardLiveState,
   listDashboardServices,
@@ -212,7 +213,7 @@ const server = http.createServer(async (req, res) => {
     if (!auth) return;
 
     if (url.pathname.startsWith('/api/')) {
-      return handleApiRequest({ req, res, url, auth });
+      return await handleApiRequest({ req, res, url, auth });
     }
 
     if (req.method === 'GET' && staticAssets.has(url.pathname)) {
@@ -318,16 +319,37 @@ async function handleApiRequest({ req, res, url, auth }) {
       });
     }
 
-    const result = await invokeControl(control, body || {});
+    const serviceDefinition = listDashboardServices(config)
-    const serviceSnapshot = await loadServiceSnapshot(
+      .find((definition) => definition.service === control.service);
-      listDashboardServices(config).find((definition) => definition.service === control.service),
+    try {
-    );
+      const result = await invokeControl(control, body || {});
-    return sendJson(res, 200, {
+      const serviceSnapshot = await loadServiceSnapshot(serviceDefinition);
-      ok: true,
+      return sendJson(res, 200, {
-      control,
+        ok: true,
-      result,
+        control,
-      service_snapshot: serviceSnapshot,
+        result,
-    });
+        service_snapshot: serviceSnapshot,
+      });
+    } catch (error) {
+      logger.warn('dashboard_control_failed', {
+        details: {
+          control,
+          error: serializeError(error),
+        },
+      });
+      const serviceSnapshot = await loadServiceSnapshot(serviceDefinition).catch((snapshotError) => ({
+        ...serviceDefinition,
+        reachable: false,
+        state: null,
+        health: null,
+        error: serializeError(snapshotError),
+      }));
+      const failure = buildDashboardControlErrorResponse(error, { control });
+      return sendJson(res, failure.statusCode, {
+        ...failure.payload,
+        service_snapshot: serviceSnapshot,
+      });
+    }
   }
 
   return sendJson(res, 404, { error: 'not_found' });

src/core/operator-dashboard.mjs

@@ -245,6 +245,21 @@ export function resolveDashboardControl({ service, action }) {
   )) || null;
 }
 
+export function buildDashboardControlErrorResponse(error, { control = null } = {}) {
+  const name = String(error?.name || 'Error');
+  const message = String(error?.message || 'Control request failed.');
+  const timedOut = name === 'TimeoutError' || /timeout|aborted/i.test(message);
+  return {
+    statusCode: timedOut ? 504 : 502,
+    payload: {
+      ok: false,
+      error: timedOut ? 'control_timeout' : 'control_failed',
+      reason: message,
+      control,
+    },
+  };
+}
+
 export function listDashboardServices(config) {
   return SERVICE_DEFINITIONS.map(([service, label, configKey]) => ({
     service,

test/operator-dashboard-app-static.test.mjs

@@ -0,0 +1,16 @@
+import test from 'node:test';
+import assert from 'node:assert/strict';
+import { readFileSync } from 'node:fs';
+
+const source = readFileSync(new URL('../src/apps/operator-dashboard.mjs', import.meta.url), 'utf8');
+
+test('operator dashboard awaits API handler promises so request errors reach the top-level catch', () => {
+  assert.equal(source.includes('return await handleApiRequest({ req, res, url, auth });'), true);
+  assert.equal(source.includes('return handleApiRequest({ req, res, url, auth });'), false);
+});
+
+test('operator dashboard control proxy catches upstream failures before sending JSON response', () => {
+  assert.match(source, /dashboard_control_failed/);
+  assert.match(source, /buildDashboardControlErrorResponse/);
+  assert.match(source, /failure.statusCode/);
+});

test/operator-dashboard.test.mjs

@@ -4,6 +4,7 @@ import assert from 'node:assert/strict';
 import {
   applyDashboardLiveEvent,
   buildDashboardBootstrap,
+  buildDashboardControlErrorResponse,
   buildLiveStatusBar,
   buildProfitabilitySummary,
   createDashboardLiveState,
@@ -104,6 +105,21 @@ test('profitability summary flags cash-flow-adjusted benchmarks after later fund
   assert.match(summary.caveats[0], /external cash flows/);
 });
 
+test('dashboard control errors become structured responses instead of uncaught failures', () => {
+  const control = resolveDashboardControl({
+    service: 'trade-executor',
+    action: 'intent-request-preflight',
+  });
+  const timeout = new DOMException('The operation was aborted due to timeout', 'TimeoutError');
+  const response = buildDashboardControlErrorResponse(timeout, { control });
+
+  assert.equal(response.statusCode, 504);
+  assert.equal(response.payload.ok, false);
+  assert.equal(response.payload.error, 'control_timeout');
+  assert.equal(response.payload.control.action, 'intent-request-preflight');
+  assert.match(response.payload.reason, /timeout/i);
+});
+
 test('control routing only resolves the allowlisted safe dashboard actions', () => {
   const refresh = resolveDashboardControl({
     service: 'liquidity-manager',