Demo Monitoring

In this lesson, you’ll learn how to install the Datadog Agent on an AWS EC2 Ubuntu/Debian instance to collect critical system metrics (CPU, memory, disk I/O, network, etc.) and then configure the Nginx integration for deep visibility into your web server’s performance.

Below is a quick reference of the main metric categories the Datadog Agent collects:

1. Install the Datadog Agent

Follow these steps to get the Agent up and running:

1. Sign up for a Datadog trial (14-day free) and skip optional steps until you reach the Agent installer.

2. Select Ubuntu/Debian as your platform.

3. Copy and run the installer command, replacing <YOUR_API_KEY> with your API key:

   DD_API_KEY="<YOUR_API_KEY>" DD_SITE="datadoghq.eu" \
     bash -c "$(curl -L https://install.datadoghq.eu/scripts/install_script_agent7.sh)"
   

   Note

   If you’re in the US region, set DD_SITE="datadoghq.com". For other regions, refer to Datadog Sites.

4. Watch the installer fetch and set up packages:

   gpg: imported: 1
   Get:1 https://apt.datadoghq.com stable Release [26.0 kB]
   Get:2 https://apt.datadoghq.com stable/7 amd64 Packages [92.1 kB]
   Fetched 120 kB in 0s (651 kB/s)
   Reading package lists...
   Installing package(s): datadog-agent datadog-signing-keys
   ...
   Unpacking datadog-agent (1:7.62.3-1) ...
   Setting up datadog-agent (1:7.62.3-1) ...
   Starting Datadog Agent
   

5. Verify the Agent status:

   sudo datadog-agent status
   

6. In the Datadog UI, click Finish, then after ~2 minutes navigate to Dashboards → Hosts to confirm your host is reporting.

You can also drill into disk I/O analytics:

2. Configure the Nginx Integration

Once the Agent is collecting system metrics, enable the Nginx integration to capture request rates, active connections, and more.

1. In the Datadog UI, go to Integrations → Integrations, search for Nginx, then click Configure.

   

2. Ensure the http_stub_status_module is compiled into your Nginx binary:

   nginx -V 2>&1 | grep -o http_stub_status_module
   

   Warning

   Exposing the stub status endpoint to the public internet is a security risk. Restrict access to localhost or your private network.

3. Update your site configuration to expose the status endpoint on localhost:80. For example, in /etc/nginx/sites-available/example.conf:

   server {
       listen 80;
       server_name localhost;
   
       access_log off;
       allow 127.0.0.1;
       deny all;
   
       location /nginx_status {
           stub_status;
           server_tokens on;
       }
   }
   

4. Reload Nginx and verify the endpoint:

   sudo nginx -t
   sudo nginx -s reload
   curl http://127.0.0.1/nginx_status
   

   Expected response:

   Active connections: 1
   server accepts handled requests
    11 11 15
   Reading: 0 Writing: 1 Waiting: 0
   

5. Back in Datadog, under Nginx Integration → Configuration, set:

   • Host: your EC2 instance
   • Endpoint: http://localhost/nginx_status

After a few minutes, open the built-in Nginx Metrics dashboard to visualize traffic, latency, error rates, and more:

For per-host details, use the Host Map:

Next Steps

With system-level metrics and Nginx integration in place, you can:

• Configure Datadog Alerts on key thresholds (e.g., high request latency).
• Build custom dashboards for error rates and traffic patterns.
• Apply APM for detailed request tracing.

Links and References

• Datadog Agent Installation
• Nginx Stub Status Module
• Datadog Nginx Integration