Explore request rate and error metrics
The IIS integration provides pre-built dashboards that help you monitor the health and performance of your web servers. In this step, you’ll become familiar with these dashboards and learn how to use them to identify and troubleshoot issues.
Did you know? If you don’t see any data, try switching the data source using the drop-down at the top of the dashboard, or verify that your IIS sites are actively receiving traffic.
IIS overview dashboard
This dashboard provides a high-level view of your IIS server health and performance.
Use this dashboard to:
- View total request throughput across all sites
- Monitor active connections and connection attempts
- Track HTTP response codes (2xx, 3xx, 4xx, 5xx) to spot error trends
- Check bytes sent and received for bandwidth analysis
Request rate and error metrics
The most critical metrics for web server monitoring are request rates and error rates.
Key metrics to watch:
- Requests per second — A sudden drop may indicate a configuration issue or upstream dependency failure. A sudden spike may signal a traffic surge or bot activity.
- HTTP 5xx error rate — Server-side errors that indicate problems with your application or server configuration. Even a small percentage of 5xx errors warrants investigation.
- HTTP 4xx error rate — Client-side errors. A rising 4xx rate might indicate broken links, API changes, or misconfigured clients.
- Request processing time — How long IIS takes to serve requests. Increasing latency is often the first sign of resource exhaustion.
Worker process metrics
Worker processes handle the actual request processing in IIS. Monitoring them helps you catch issues before they affect users.
Key metrics to watch:
- Worker process CPU usage — High CPU in a worker process may indicate an infinite loop, expensive computation, or resource contention.
- Worker process memory — Steadily increasing memory usage often indicates a memory leak. Worker process recycling resets memory but interrupts active requests.
- Active requests per worker — Helps you understand load distribution across application pools.
Connection metrics
Connection metrics help you understand capacity utilization.
Key metrics to watch:
- Current connections — The number of active client connections. Compare against your server’s connection limits.
- Connection attempts per second — A spike in connection attempts without a corresponding rise in active connections may indicate connection timeouts or refused connections.
- Maximum connections — Helps you plan capacity and set appropriate connection limits.