feat(health): add multi-service health check endpoint

backend/api/src/index.js

@@ -19,6 +19,7 @@ import profileRoutes from './routes/profileRoutes.js';
 import loadRoutes from './routes/loadRoutes.js';
 import truckRoutes from './routes/truckRoutes.js';
 import authRoutes from './routes/authRoutes.js';
+import healthRoutes from './routes/healthRoutes.js';
 
 import logger from './middleware/logger.js';
 import { requestIdMiddleware, requestLogger } from './middleware/requestId.js';
@@ -172,18 +173,7 @@ app.use('/api/v1/trips', tripRoutes);
 // ============================================================================
 // REST API ROUTING
 // ============================================================================
-app.get('/api/health', (req, res) => {
-  res.json({
-    status: 'healthy',
-    timestamp: new Date(),
-    service: 'Truxify API',
-    uptime: process.uptime(),
-    env: {
-      bypass_auth: process.env.BYPASS_AUTH === 'true',
-      node_version: process.version
-    }
-  });
-});
+app.use('/api/health', healthRoutes);
 
 app.use('/api/orders', orderRoutes);
 app.use('/api/driver', driverRoutes);

backend/api/src/routes/healthRoutes.js

@@ -0,0 +1,97 @@
+import express from 'express';
+import { supabase, mongoDb, redisClient, firebaseAdmin } from '../config/db.js';
+
+const router = express.Router();
+
+const DEFAULT_TIMEOUT_MS = 400;
+const _parsedTimeout = Number(process.env.HEALTHCHECK_TIMEOUT_MS);
+const CHECK_TIMEOUT_MS =
+  Number.isFinite(_parsedTimeout) && _parsedTimeout > 0 ? _parsedTimeout : DEFAULT_TIMEOUT_MS;
+
+function withTimeout(promise) {
+  let timer;
+  return Promise.race([
+    promise,
+    new Promise((_, reject) => {
+      timer = setTimeout(() => reject(new Error('healthcheck timeout')), CHECK_TIMEOUT_MS);
+    }),
+  ]).finally(() => clearTimeout(timer));
+}
+
+async function checkSupabase() {
+  if (!supabase) return 'not_configured';
+  try {
+    const { error } = await withTimeout(
+      supabase.from('profiles').select('id').limit(1)
+    );
+    return error ? 'failed' : 'connected';
+  } catch {
+    return 'failed';
+  }
+}
+
+async function checkMongo() {
+  if (!mongoDb) return 'not_configured';
+  try {
+    await withTimeout(mongoDb.admin().ping());
+    return 'connected';
+  } catch {
+    return 'failed';
+  }
+}
+
+async function checkRedis() {
+  if (!redisClient) return 'not_configured';
+  try {
+    const reply = await withTimeout(redisClient.ping());
+    return reply === 'PONG' ? 'connected' : 'failed';
+  } catch {
+    return 'failed';
+  }
+}
+
+function checkFirebase() {
+  return firebaseAdmin ? 'configured' : 'not_configured';
+}
+
+function checkPolygon() {
+  return process.env.POLYGON_RPC_URL ? 'configured' : 'not_configured';
+}
+
+const CRITICAL_UNHEALTHY = new Set(['failed', 'not_configured']);
+
+// GET /api/health — full dependency check; returns 503 when a critical service fails
+router.get('/', async (req, res) => {
+  const [supabaseStatus, mongoStatus, redisStatus] = await Promise.all([
+    checkSupabase(),
+    checkMongo(),
+    checkRedis(),
+  ]);
+
+  const services = {
+    supabase: supabaseStatus,
+    mongodb: mongoStatus,
+    redis: redisStatus,
+    firebase: checkFirebase(),
+    polygon: checkPolygon(),
+  };
+
+  const criticalFailed =
+    CRITICAL_UNHEALTHY.has(supabaseStatus) || CRITICAL_UNHEALTHY.has(mongoStatus);
+
+  const status = criticalFailed ? 'degraded' : 'ok';
+  const httpStatus = criticalFailed ? 503 : 200;
+
+  return res.status(httpStatus).json({
+    status,
+    services,
+    uptime: process.uptime(),
+  });
+});
+
+// GET /api/health/live — liveness probe; always 200 as long as the process is up
+router.get('/live', (req, res) => {
+  res.json({ status: 'ok', uptime: process.uptime() });
+});
+
+export default router;

docker-compose.yml

@@ -18,6 +18,12 @@ services:
       - mongo
       - redis
       - db
+    healthcheck:
+      test: ["CMD-SHELL", "wget -qO- http://localhost:5000/api/health/live || curl -sf http://localhost:5000/api/health/live"]
+      interval: 30s
+      timeout: 5s
+      retries: 5
+      start_period: 15s
 
   ml-engine:
     build: