The 3 monitoring tools I use for every production app

A production outage at 2 AM. The customer calls before you know there is a problem. This is what I set up to prevent that.

1. Uptime Monitoring

Question: Is my site up?

Tool: UptimeRobot or Checkly (free tier is enough to start)

Setup: Check every 5 minutes from multiple regions

-- Alert if:
- Response time > 3 seconds
- Status code != 200
- SSL certificate expires in 30 days

2. Error Tracking

Question: What exceptions are happening?

Tool: Sentry (free tier is generous)

Setup: Capture all unhandled exceptions and promise rejections

// In your app
import * as Sentry from '@sentry/react';

Sentry.init({
dsn: 'your-dsn',
environment: process.env.NODE_ENV,
});

window.addEventListener('unhandledrejection', (event) => {
Sentry.captureException(event.reason);
});

3. Logging + Metrics

Question: What is happening right now?

Tool: DataDog, Grafana+Prometheus, or CloudWatch

The key metrics to watch:

• Error rate (errors per minute)
• P99 latency
• CPU/Memory usage

The Alerting Rule

Only alert on things that matter:

• Site is down
• Error rate spiked > 5%
• P99 latency > 5 seconds

Don’t alert on every error. You will ignore alerts.

One More Thing

Have a status page ready. When things go wrong, update it before customers complain.