Website Uptime Monitoring: How to Know When Your Site Goes Down
Uptime monitoring continuously checks your website and alerts you the moment it becomes unreachable. Here is how it works, what downtime actually costs, and how to set it up.
Website Uptime Monitoring: How to Know When Your Site Goes Down
Uptime monitoring makes periodic HTTP requests to your website and alerts you when it stops responding. Without monitoring, you find out about downtime from angry customers or by accidentally visiting your own site.
What Uptime Monitoring Checks
Each check makes an HTTP GET request to your domain's HTTPS endpoint with a 10-second timeout. It records:
- HTTP status code — 200 = up, 5xx = server error, 4xx = client error, timeout = down
- Response time — milliseconds from request to first byte
- Final URL — where the domain ends up after redirects
- Server header — web server software identity
- HTTPS enforcement — whether HTTP redirects to HTTPS
A domain is classified as down if:
- Connection times out (10 seconds with no response)
- DNS resolution fails
- SSL handshake fails
- Server returns 5xx status code
What Causes Website Downtime
Server Issues
- Server crashed or ran out of memory
- Disk full (web server logs filling disk)
- CPU overload from traffic spike or runaway process
DNS Failures
- Domain expired → DNS stops resolving
- Nameserver changed but old one decommissioned
- DNS provider outage
SSL Certificate Expiry
An expired SSL certificate causes connection failures in all modern browsers. Users get a security error before the page loads.
Hosting Provider Outage
Your entire hosting provider or datacenter goes offline.
DDoS Attack
Distributed Denial of Service attack overwhelms server resources.
Accidental Misconfiguration
A deployment breaks the app, a config change takes the server offline.
How to Set Up Uptime Monitoring
Uptime checks run automatically as part of every domain scan. To get immediate alerts:
- Go to domain detail → Alerts → Create Alert Rule.
- Set trigger: Uptime Down.
- Choose notification channel: Slack or Webhook (for immediate response), plus Email (for documentation).
- Optionally add: Uptime Up alert (recovery notification).
- Save.
Now whenever a scan detects the domain is unreachable, the alert fires immediately.
Scan Frequency and Alert Speed
How quickly you detect downtime depends on scan frequency:
| Scan Interval | Max Detection Delay |
|---|---|
| Every 6 hours | Up to 6 hours |
| Every 24 hours | Up to 24 hours |
| Weekly | Up to 7 days |
For production domains, every 6 hours is the recommended minimum. For mission-critical uptime where you need alerts within minutes, pair ElasticDomain's monitoring with a dedicated uptime service that checks every 1-5 minutes.
Response Time Benchmarks
| Response Time | Rating |
|---|---|
| Under 200ms | Excellent |
| 200-500ms | Good |
| 500ms-1s | Acceptable |
| Over 1s | Slow — investigate |
| Over 3s | Very slow — likely affecting user experience and bounce rate |
What to Do When Downtime Is Detected
- Acknowledge the alert — confirm the domain is actually down (visit in your browser).
- Check your server — SSH in and check if the web server process is running.
- Check DNS — confirm the domain is still resolving to the right IP.
- Check SSL — confirm the certificate is not expired.
- Check your hosting provider status — if it is a provider outage, you are waiting for them.
- Review recent deployments — did a recent code or config change cause the issue?
Uptime History
Every uptime check is stored in scan history. Over time this builds a record of:
- Your domain's historical availability
- Response time trends
- When outages occurred and how long they lasted
This data is valuable for client SLA reporting and for understanding patterns (e.g., a daily spike in response time that correlates with a scheduled backup job).