Monitoring Setup - Observability Configuration

Overview

Sets up the "detect before users complain" system. Minimum viable monitoring that grows with your startup.

---

1. When to Apply

Trigger	Behavior
Post-launch monitoring setup	Full observability stack
"monitoring", "alerting"	Tool-specific configuration
First production incident	Retroactive monitoring gaps

---

2. Monitoring Stack by Budget

Budget	Stack	Monthly Cost
$0	Sentry (free) + UptimeRobot + CloudWatch basic	$0
$0-50	Sentry + Grafana Cloud (free) + Better Uptime	$0-50
$50-500	Datadog (Pro) or Grafana Cloud (Pro)	$50-500

---

3. Essential Alerts (Day 1)

## Must-Have Alerts

| Alert | Condition | Channel | Priority |
|-------|-----------|---------|----------|
| **Site Down** | Health check fails 3x | Slack + PagerDuty | P0 |
| **Error Rate Spike** | >5% 5xx in 5 min | Slack | P1 |
| **Slow Response** | p95 > 2s for 5 min | Slack | P2 |
| **DB Connections** | >80% pool used | Slack | P1 |
| **Disk Space** | >85% used | Slack | P2 |
| **Memory** | >90% for 10 min | Slack | P1 |
| **SSL Expiry** | <14 days to expiry | Email | P2 |

---

4. Quick Setup: Sentry (Error Tracking)

// Next.js — sentry.client.config.ts
import * as Sentry from "@sentry/nextjs"

Sentry.init({
  dsn: process.env.NEXT_PUBLIC_SENTRY_DSN,
  tracesSampleRate: 0.1,      // 10% of transactions
  replaysSessionSampleRate: 0, // Save cost, enable when needed
  environment: process.env.NODE_ENV,
})

# FastAPI — main.py
import sentry_sdk
from sentry_sdk.integrations.fastapi import FastApiIntegration

sentry_sdk.init(
    dsn=os.getenv("SENTRY_DSN"),
    traces_sample_rate=0.1,
    environment=os.getenv("ENVIRONMENT", "production"),
    integrations=[FastApiIntegration()],
)

---

5. Health Check Endpoint

// app/api/health/route.ts
export async function GET() {
  const checks = {
    status: 'ok',
    timestamp: new Date().toISOString(),
    uptime: process.uptime(),
    checks: {
      database: await checkDB(),
      redis: await checkRedis(),
    }
  }
  const allHealthy = Object.values(checks.checks).every(c => c === 'ok')
  return Response.json(checks, { status: allHealthy ? 200 : 503 })
}

---

6. Tool Coordination

Tool	Purpose
Write	Generate config files, health endpoints
Read	Analyze existing monitoring setup
Bash	Install SDKs (npm install @sentry/nextjs)

---

7. Boundaries

Will:

• Generate monitoring configurations for major platforms
• Set up error tracking (Sentry), uptime monitoring, basic APM
• Define alert thresholds appropriate for startups
• Create health check endpoints

Will Not:

• Create cloud provider accounts
• Set up PagerDuty/OpsGenie on-call rotations
• Build custom dashboards (recommend tools instead)
• Monitor business metrics (use metrics-tracker)