Observability

Monitor your Tusflow API health and performance with built-in observability features

Overview

Tusflow provides comprehensive observability through health checks and integrations with monitoring services like BetterStack and OpenStatus. Monitor your API's health, track performance metrics, and get alerted about issues.

Health Check Endpoint

The /health endpoint provides real-time status of all critical services:

GET /health
 
// Response 200 OK
{
  "status": "ok"
}
 
// Response 503 Service Unavailable
{
  "status": "error"
}
 
Headers:
  X-Health-Check: ok | error
  X-Response-Time: latency_ms
  User-Agent: OpenStatus/1.0

The health check:

Verifies S3 connectivity
Validates Redis connection
Measures response latency
Returns detailed status headers

BetterStack Integration

Monitor your Tusflow API with BetterStack's comprehensive observability platform:

Configure BetterStack

Create a BetterStack account
Add a new monitor in BetterStack dashboard
Select "HTTP" monitor type
Enter your Tusflow health endpoint URL

Set Up Alerts

Configure notification channels (email, Slack, etc.)
Set alert thresholds
Define escalation policies
Add team members

Monitor Metrics

Response times
Error rates
Availability percentage
Service dependencies

Example BetterStack configuration:

// BetterStack Monitor Settings
{
  "name": "Tusflow API Health",
  "url": "https://api.Tusflow.dev/health",
  "check_frequency": 30,
  "regions": ["us", "eu", "asia"],
  "alert_threshold": {
    "response_time": 1000,
    "status_code": [200]
  }
}

OpenStatus Integration

Incident

Monitor your API status with OpenStatus's synthetic monitoring:

Create Monitor

Sign up for OpenStatus
Create a new HTTP monitor
Configure health check endpoint
Set check frequency

Configure Assertions

Status code should be 200
Response should contain {"status": "ok"}
Response time under threshold
Headers present and valid

Set Up Status Page

Create public status page
Add monitor to status page
Configure incident alerts
Share status page URL

Example OpenStatus configuration:

// OpenStatus Monitor Settings
{
  "name": "Tusflow API",
  "url": "https://api.Tusflow.dev/health",
  "method": "GET",
  "frequency": "1m",
  "regions": ["iad", "sfo", "fra"],
  "assertions": [
    {
      "type": "statusCode",
      "value": 200
    },
    {
      "type": "body",
      "value": "status",
      "operator": "contains"
    }
  ]
}

Implementation Details

The health check endpoint implementation:

app.get('/health', async (c) => {
    const startTime = performance.now();
 
    try {
        // Initialize clients
        const s3Client = new S3Client({
            region: c.env.AWS_REGION,
            endpoint: c.env.AWS_ENDPOINT,
            credentials: {
                accessKeyId: c.env.AWS_ACCESS_KEY_ID,
                secretAccessKey: c.env.AWS_SECRET_ACCESS_KEY,
            },
            forcePathStyle: true,
        });
 
        const redis = new Redis({
            url: c.env.UPSTASH_REDIS_REST_URL,
            token: c.env.UPSTASH_REDIS_REST_TOKEN,
        });
 
        // Check services health in parallel
        const [s3Health, redisHealth] = await Promise.all([
            checkS3Health(s3Client),
            checkRedisHealth(redis),
        ]);
 
        const latency = Math.round(performance.now() - startTime);
 
        // Set response headers
        c.header('X-Health-Check', s3Health && redisHealth ? 'ok' : 'error');
        c.header('X-Response-Time', `${latency}ms`);
        c.header('User-Agent', 'OpenStatus/1.0');
 
        return c.json({ 
            status: s3Health && redisHealth ? 'ok' : 'error' 
        }, s3Health && redisHealth ? 200 : 503);
    } catch (error) {
        c.header('X-Health-Check', 'error');
        c.header('X-Response-Time', '0ms');
        return c.json({ status: 'error' }, 503);
    }
});

Best Practices

Monitoring Setup
- Monitor from multiple regions
- Set appropriate check frequencies
- Configure meaningful thresholds
- Enable incident alerts
Health Checks
- Keep checks lightweight
- Include critical dependencies
- Set reasonable timeouts
- Monitor response times
Alerting
- Define clear escalation policies
- Set up redundant notifications
- Configure maintenance windows
- Document incident responses

Regular health checks and monitoring are essential for maintaining high availability and quickly responding to issues.

On this page