How Do I Monitor My DigitalOcean Droplet?

Built-In Monitoring

DigitalOcean includes free resource monitoring for every Droplet. Enable it during creation or afterward on the Droplet's Graphs tab. You get 14-day history of CPU, memory, disk I/O, disk usage, bandwidth, and load average — all without installing anything.

Step 1 — Enable Monitoring

If not already on, go to your Droplet → Graphs → click Enable Monitoring. A lightweight agent installs automatically. Graphs populate within a few minutes.

Step 2 — Set Up Alert Policies

Go to Monitoring → Alert Policies → Create Alert Policy. Useful defaults:

• CPU usage above 80% for 5 minutes
• Memory usage above 90% for 5 minutes
• Disk usage above 85%
• Droplet down — no heartbeat received

Alerts can notify you via email or Slack webhook so you catch problems before users do.

Step 3 — Monitor from the Command Line

SSH into your Droplet for real-time diagnostics:

htop                    # interactive CPU/memory/process view
df -h                   # disk space
free -h                 # memory usage
journalctl -f           # live system logs
pm2 monit               # if using PM2 for Node.js

Step 4 — Application-Level Monitoring

System metrics tell you that something is wrong; app metrics tell you what. Add health checks and uptime monitoring:

• UptimeRobot or Better Stack — ping your URL every minute, alert on downtime
• PM2 — pm2 monit for Node.js process metrics
• Docker — docker stats for container resource usage

Step 5 — Log Management

Centralize logs so you can search across restarts:

# Nginx access and error logs
tail -f /var/log/nginx/access.log
tail -f /var/log/nginx/error.log

# System logs
journalctl -u nginx --since "1 hour ago"

For production, ship logs to a service like Better Stack, Datadog, or a self-hosted Loki instance.

Monitoring Checklist

• Enable DigitalOcean monitoring on every Droplet
• Set alert policies for CPU, memory, and disk thresholds
• Add external uptime checks for your public URLs
• Review graphs weekly to spot gradual resource creep

Good monitoring turns "my site is down" into "CPU spiked at 2:14 AM — here is why."