Logging e metriche di Nginx

Nginx logs are vital for understanding your server's activity and health. They record every request and any issues that occur.

Think of them as your server's diary. By reviewing these logs, you can monitor performance, troubleshoot errors, identify security threats, and gather valuable usage metrics.

Access logs record every request Nginx processes. Each line details a single client request, providing insights into who accessed what, and when.

• Client IP: Who made the request.
• Request method & URL: What was requested.
• Status code: Was it successful (200), an error (404, 500), etc.?
• Response size: How much data was sent back.
• User-Agent: The client's browser/device.

By default, Nginx uses the 'combined' log format, which includes standard information. You'll typically find your access logs in /var/log/nginx/access.log.

Here's what a common entry looks like:

Nginx allows you to define custom log formats using the log_format directive. This is powerful for tailoring logs to your specific monitoring or analysis needs.

You can include variables like $remote_addr (client IP), $request (full request line), $status (response status code), $body_bytes_sent, and many more.

Let's create a custom log format named 'my_json_format' that outputs logs in JSON. This is great for automated parsing!

Place this inside your http block in nginx.conf:

Error logs record diagnostic information about issues Nginx encounters, from configuration problems to backend connection failures.

They are crucial for debugging and maintaining server stability. You'll typically find them at /var/log/nginx/error.log.

The error_log directive allows you to specify the log file and the severity level. Nginx logs messages at or above the specified level.

• debug: Most verbose, for deep debugging.
• info: Informational messages.
• notice: Normal but significant conditions.
• warn: Potentially problematic situations.
• error: An error occurred.
• crit: Critical conditions.
• alert: Action must be taken immediately.
• emerg: System is unusable.

You can configure the error log level globally in the main or http block, or specifically within a server or location block.

For production, warn or error is common to keep logs manageable. For debugging, you might temporarily set it to info or debug.

In production, logs can grow very large, very quickly. Log rotation is the process of archiving, compressing, and deleting old log files to prevent them from filling up your disk space.

While Nginx doesn't handle rotation itself, tools like logrotate (common on Linux) are used to manage this. You configure logrotate to periodically rotate Nginx's log files.

You've learned that Nginx access logs track client requests, while error logs report server issues. Customizing log formats with log_format allows you to tailor output for specific metrics or tools.

Understanding log levels (e.g., warn, error) helps control verbosity, and knowing about log rotation ensures disk space management.

These logging fundamentals are essential for effective monitoring and troubleshooting of your Nginx server.