---
name: nginx-request-logging
description: Guide for configuring Nginx web server with custom request logging, rate limiting, and error pages. This skill should be used when tasks involve Nginx installation, configuration, custom log formats, rate limiting setup, or custom error page creation.
---

# Nginx Request Logging Configuration

This skill provides guidance for configuring Nginx web servers with custom logging, rate limiting, and error handling.

## When to Use This Skill

Apply this skill when tasks involve:
- Installing and configuring Nginx
- Setting up custom log formats
- Implementing rate limiting
- Creating custom error pages (404, 500, etc.)
- Configuring Nginx to listen on non-standard ports

## Pre-Configuration Analysis

Before modifying any Nginx configuration:

1. **Examine existing configuration structure**
   - Read `/etc/nginx/nginx.conf` to understand the current setup
   - Check for existing `include` directives to understand file organization
   - Identify where log formats, rate limiting zones, and other global settings are defined

2. **Check system state**
   - Verify if Nginx is already installed: `which nginx` or `nginx -v`
   - Check if Nginx is already running: `pgrep nginx` or `ps aux | grep nginx`
   - Verify if the target port is available: `ss -tlnp | grep <port>` or `netstat -tlnp | grep <port>`

3. **Backup original configuration**
   - Create a backup before modifications: `cp /etc/nginx/nginx.conf /etc/nginx/nginx.conf.bak`

## Configuration Approach

### Directory Structure

Nginx configurations typically follow this hierarchy:
- `/etc/nginx/nginx.conf` - Main configuration (global settings, log formats, rate limiting zones)
- `/etc/nginx/conf.d/` - Site-specific configurations (server blocks)
- `/etc/nginx/sites-available/` and `/etc/nginx/sites-enabled/` - Alternative site management (Debian-based)

### Configuration Placement Guidelines

| Setting Type | Location | Reason |
|-------------|----------|--------|
| Log format definitions | `nginx.conf` (http block) | Must be defined before use in server blocks |
| Rate limiting zones | `nginx.conf` (http block) | Zones are shared across server blocks |
| Server blocks | `conf.d/*.conf` | Modular, easy to manage |
| Custom error pages | Server block or location block | Context-specific |

### Rate Limiting Configuration

Rate limiting requires two parts:

1. **Zone definition** (in http block of nginx.conf):
   ```nginx
   limit_req_zone $binary_remote_addr zone=zonename:10m rate=10r/s;
   ```

2. **Zone application** (in server or location block):
   ```nginx
   limit_req zone=zonename burst=5 nodelay;
   ```

### Custom Log Format

Define custom log formats in the http block:
```nginx
log_format custom_format '$remote_addr - $remote_user [$time_local] '
                         '"$request" $status $body_bytes_sent '
                         '"$http_referer" "$http_user_agent"';
```

Apply in server block:
```nginx
access_log /var/log/nginx/custom_access.log custom_format;
```

## Service Management

Nginx service management varies by environment:

| Environment | Start Command | Reload Command | Stop Command |
|-------------|--------------|----------------|--------------|
| systemd | `systemctl start nginx` | `systemctl reload nginx` | `systemctl stop nginx` |
| Direct | `nginx` | `nginx -s reload` | `nginx -s stop` |
| Docker/Container | `nginx -g 'daemon off;'` | `nginx -s reload` | `nginx -s quit` |

**Important**: Always test configuration before starting/reloading:
```bash
nginx -t
```

## Verification Strategies

### Basic Functionality
```bash
curl -s http://localhost:<port>/
curl -s -o /dev/null -w "%{http_code}" http://localhost:<port>/nonexistent
```

### Rate Limiting Verification

Rate limiting requires **concurrent** requests to trigger. Sequential requests will not exceed the rate limit.

**Correct approach** (parallel requests):
```bash
seq 20 | xargs -P 20 -I {} curl -s -o /dev/null -w "%{http_code}\n" http://localhost:<port>/
```

**Incorrect approach** (will not trigger rate limiting):
```bash
for i in {1..20}; do curl -s http://localhost:<port>/; done  # Too slow, sequential
```

### Log Verification
```bash
tail -f /var/log/nginx/access.log
tail -f /var/log/nginx/error.log
```

## Common Pitfalls

1. **Log format not found**: Log format must be defined in nginx.conf before being referenced in server blocks

2. **Rate limiting not triggering**: Sequential requests are too slow; use parallel requests with `xargs -P` or similar

3. **Configuration syntax errors**: Always run `nginx -t` before starting or reloading

4. **Port already in use**: Check with `ss -tlnp` before configuring a new port

5. **systemctl not available**: In containers or minimal environments, use `nginx` command directly

6. **Default site conflicts**: Remove or disable default site configuration when creating custom configurations:
   ```bash
   rm -f /etc/nginx/sites-enabled/default
   ```

7. **Missing directories**: Verify required directories exist before writing configuration:
   ```bash
   ls -la /etc/nginx/conf.d/
   ```

## Execution Efficiency

- **Batch file operations**: Create multiple static files (index.html, 404.html, etc.) in parallel when possible
- **Combine verification steps**: Test multiple endpoints in a single verification pass
- **Plan verification upfront**: Determine the testing strategy before implementation
- **Use idempotent commands**: Prefer `mkdir -p`, `rm -f` to handle existing/missing files gracefully

## Example Workflow

1. Check system state (Nginx installed, running, port availability)
2. Read existing nginx.conf structure
3. Backup configuration
4. Create required directories and static content
5. Modify nginx.conf for global settings (log format, rate limiting zone)
6. Create server configuration in conf.d/
7. Remove conflicting default configurations
8. Test configuration with `nginx -t`
9. Start/reload Nginx service
10. Verify all functionality (main page, error pages, rate limiting, logs)