295 lines
6.5 KiB
Markdown
295 lines
6.5 KiB
Markdown
# Nginx Role
|
|
|
|
This Ansible role installs and configures Nginx as a reverse proxy for web applications.
|
|
|
|
## Features
|
|
|
|
- Installs Nginx
|
|
- Configurable worker processes and connections
|
|
- Gzip compression support
|
|
- SSL/TLS configuration
|
|
- Modular vhost configuration via `/etc/nginx/conf.d/`
|
|
- Zero-downtime reloads
|
|
- Configurable logging backend (journald or traditional files)
|
|
- Automatic logrotate configuration for file-based logging
|
|
|
|
## Requirements
|
|
|
|
- Systemd-based Linux distribution
|
|
- Root/sudo access
|
|
|
|
## Role Variables
|
|
|
|
See `defaults/main.yml` for all available variables and their default values.
|
|
|
|
### Key Configuration
|
|
|
|
The role provides sensible defaults for worker processes, connection limits, upload sizes, compression, and SSL/TLS settings. Override as needed in your inventory.
|
|
|
|
### Logging Configuration
|
|
|
|
```yaml
|
|
# Logging backend: 'journald' (systemd journal) or 'file' (traditional logs)
|
|
nginx_log_backend: journald # Default: journald
|
|
|
|
# Logrotate settings (only used when nginx_log_backend: file)
|
|
nginx_logrotate_rotate: 14 # Keep 14 days of logs
|
|
nginx_logrotate_frequency: daily # daily|weekly|monthly
|
|
nginx_logrotate_compress: true # Compress rotated logs
|
|
```
|
|
|
|
**journald backend (default):**
|
|
- Logs sent to systemd journal via syslog
|
|
- Centralized with other system logs
|
|
- Managed by systemd-journald (size limits, retention, compression)
|
|
- View with: `journalctl -u nginx`
|
|
|
|
**file backend:**
|
|
- Traditional `/var/log/nginx/*.log` files
|
|
- Automatic logrotate configuration deployed
|
|
- Useful for external log aggregation tools
|
|
|
|
## Dependencies
|
|
|
|
None.
|
|
|
|
## Example Playbook
|
|
|
|
### Basic Installation
|
|
|
|
```yaml
|
|
---
|
|
- hosts: servers
|
|
become: true
|
|
roles:
|
|
- role: nginx
|
|
```
|
|
|
|
### Custom Configuration
|
|
|
|
```yaml
|
|
---
|
|
- hosts: servers
|
|
become: true
|
|
roles:
|
|
- role: nginx
|
|
vars:
|
|
nginx_worker_processes: 4
|
|
nginx_worker_connections: 2048
|
|
nginx_client_max_body_size: 500M
|
|
```
|
|
|
|
## Service Management
|
|
|
|
The role creates handlers for managing nginx:
|
|
|
|
```yaml
|
|
notify: Reload nginx # Graceful reload (zero downtime)
|
|
notify: Restart nginx # Full restart
|
|
```
|
|
|
|
## Vhost Configuration Pattern
|
|
|
|
This role is designed to work with service-specific vhost configurations. Each service role should:
|
|
|
|
1. Deploy its vhost config to `/etc/nginx/conf.d/<service>.conf`
|
|
2. Notify the nginx reload handler
|
|
3. Use a variable to enable/disable nginx integration
|
|
|
|
### Example Service Integration
|
|
|
|
In your service role (e.g., `immich`):
|
|
|
|
**defaults/main.yml:**
|
|
```yaml
|
|
immich_nginx_enabled: false
|
|
immich_nginx_hostname: immich.example.com
|
|
```
|
|
|
|
**tasks/main.yml:**
|
|
```yaml
|
|
- name: Deploy nginx vhost for service
|
|
ansible.builtin.template:
|
|
src: nginx-vhost.conf.j2
|
|
dest: /etc/nginx/conf.d/myservice.conf
|
|
validate: nginx -t
|
|
when: myservice_nginx_enabled
|
|
notify: Reload nginx
|
|
|
|
- name: Remove nginx vhost when disabled
|
|
ansible.builtin.file:
|
|
path: /etc/nginx/conf.d/myservice.conf
|
|
state: absent
|
|
when: not myservice_nginx_enabled
|
|
notify: Reload nginx
|
|
```
|
|
|
|
**templates/nginx-vhost.conf.j2:**
|
|
```nginx
|
|
server {
|
|
listen 80;
|
|
server_name {{ myservice_nginx_hostname }};
|
|
|
|
location / {
|
|
proxy_pass http://127.0.0.1:{{ myservice_port }};
|
|
proxy_set_header Host $http_host;
|
|
proxy_set_header X-Real-IP $remote_addr;
|
|
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
|
|
proxy_set_header X-Forwarded-Proto $scheme;
|
|
}
|
|
}
|
|
```
|
|
|
|
**handlers/main.yml:**
|
|
```yaml
|
|
- name: Reload nginx
|
|
ansible.builtin.systemd:
|
|
name: nginx
|
|
state: reloaded
|
|
```
|
|
|
|
## Independent Deployments
|
|
|
|
This pattern allows for independent service deployments:
|
|
|
|
1. **Deploy service A** → Only touches `/etc/nginx/conf.d/serviceA.conf` → Reload nginx
|
|
2. **Deploy service B** → Only touches `/etc/nginx/conf.d/serviceB.conf` → Reload nginx
|
|
3. **No downtime** for other services during deployment
|
|
|
|
## Log Management
|
|
|
|
### Journald Backend (Default)
|
|
|
|
When `nginx_log_backend: journald`, logs are sent to systemd journal:
|
|
|
|
```bash
|
|
# View all nginx logs
|
|
journalctl -u nginx -f
|
|
|
|
# Last 100 lines
|
|
journalctl -u nginx -n 100
|
|
|
|
# Filter by priority (error, warning, info)
|
|
journalctl -u nginx -p err
|
|
|
|
# Time range
|
|
journalctl -u nginx --since "1 hour ago"
|
|
|
|
# Export to file
|
|
journalctl -u nginx > nginx-logs.txt
|
|
```
|
|
|
|
**Benefits:**
|
|
- Centralized with all system logs
|
|
- Automatic rotation/compression via systemd-journald
|
|
- Structured metadata (timestamps, priorities)
|
|
- No separate logrotate configuration needed
|
|
|
|
### File Backend
|
|
|
|
When `nginx_log_backend: file`, logs are written to:
|
|
- `/var/log/nginx/access.log` - Access logs
|
|
- `/var/log/nginx/error.log` - Error logs
|
|
|
|
```bash
|
|
# View traditional log files
|
|
tail -f /var/log/nginx/access.log
|
|
tail -f /var/log/nginx/error.log
|
|
```
|
|
|
|
Logrotate is automatically configured to:
|
|
- Rotate daily (configurable)
|
|
- Keep 14 days (configurable)
|
|
- Compress old logs
|
|
- Reload nginx gracefully after rotation
|
|
|
|
### Switching Backends
|
|
|
|
To switch from journald to file logging:
|
|
|
|
```yaml
|
|
- hosts: servers
|
|
roles:
|
|
- role: nginx
|
|
vars:
|
|
nginx_log_backend: file
|
|
nginx_logrotate_rotate: 30 # Keep 30 days
|
|
```
|
|
|
|
To switch back to journald:
|
|
|
|
```yaml
|
|
- hosts: servers
|
|
roles:
|
|
- role: nginx
|
|
vars:
|
|
nginx_log_backend: journald
|
|
```
|
|
|
|
The role automatically removes logrotate config when using journald.
|
|
|
|
## Configuration Validation
|
|
|
|
The role automatically validates nginx configuration before applying changes using `nginx -t`.
|
|
|
|
Manual validation:
|
|
```bash
|
|
nginx -t # Test configuration
|
|
nginx -t -c /path/to/conf # Test specific config file
|
|
```
|
|
|
|
## Troubleshooting
|
|
|
|
### Check nginx status
|
|
```bash
|
|
systemctl status nginx
|
|
```
|
|
|
|
### Test configuration
|
|
```bash
|
|
nginx -t
|
|
```
|
|
|
|
### Reload configuration
|
|
```bash
|
|
systemctl reload nginx
|
|
```
|
|
|
|
### View error logs
|
|
```bash
|
|
journalctl -u nginx -n 100
|
|
# or
|
|
tail -f /var/log/nginx/error.log
|
|
```
|
|
|
|
### List loaded vhost configs
|
|
```bash
|
|
ls -la /etc/nginx/conf.d/
|
|
```
|
|
|
|
## SSL/TLS Support
|
|
|
|
For SSL support, you can:
|
|
|
|
1. **Manual certificates:** Place certs in `/etc/ssl/` and reference in vhost configs
|
|
2. **Let's Encrypt:** Use certbot or similar tools (can be added to playbook)
|
|
3. **Self-signed:** Generate with `openssl` for testing
|
|
|
|
The base nginx.conf includes SSL protocol configuration that applies to all vhosts.
|
|
|
|
## Performance Tuning
|
|
|
|
Adjust these variables based on your workload:
|
|
|
|
- `nginx_worker_processes`: Set to number of CPU cores
|
|
- `nginx_worker_connections`: Increase for high traffic (check `ulimit -n`)
|
|
- `nginx_client_max_body_size`: Increase for large file uploads
|
|
|
|
## License
|
|
|
|
MIT
|
|
|
|
## Author Information
|
|
|
|
Created for managing reverse proxy configurations in NAS/homelab environments.
|