Hey there! Let's break down your Nginx questions clearly:

The exact command depends on your operating system's init system:

• Systemd-based systems (Ubuntu 16.04+, Debian 9+, CentOS 7+):
  Use systemctl to start Nginx and verify it's running:

  sudo systemctl start nginx
  # 检查启动状态
  sudo systemctl status nginx
  

  If you want Nginx to start automatically on system boot, run:

  sudo systemctl enable nginx
  

• SysVinit-based systems (older Ubuntu/Debian/CentOS versions):
  Use the service command:

  sudo service nginx start
  # 设置开机自启
  sudo update-rc.d nginx defaults
  

• Direct binary execution (if init system commands fail):
  Locate your Nginx binary (usually /usr/sbin/nginx) and run:

  sudo /usr/sbin/nginx
  

  If you get an error about a locked PID file, you might need to remove it first:

  sudo rm -f /var/run/nginx.pid
  sudo /usr/sbin/nginx
  

There are a few reliable ways to keep Nginx running automatically:

Method 1: Use systemd's built-in restart feature #

Systemd has native support for restarting failed services. Here's how to set it up:

1. Edit the Nginx service override file:

   sudo systemctl edit nginx
   

2. Paste the following content into the editor:

   [Service]
   Restart=always
   RestartSec=5
   

   • Restart=always tells systemd to restart Nginx no matter why it stopped (normal exit, crash, etc.)
   • RestartSec=5 adds a 5-second delay before restarting to avoid loop issues
3. Save and exit the editor, then reload systemd and restart Nginx:

   sudo systemctl daemon-reload
   sudo systemctl restart nginx
   

Method 2: Cron job to check and restart Nginx #

If you prefer a simpler approach without modifying systemd, create a cron job that checks Nginx's status periodically:

1. Create a shell script (e.g., /usr/local/bin/check_nginx.sh) with this content:

   #!/bin/bash
   # Check if Nginx is running
   if ! pgrep nginx > /dev/null; then
       sudo systemctl start nginx
       # Optional: Log the restart
       echo "$(date): Nginx was not running, restarted it" >> /var/log/nginx_restart.log
   fi
   

2. Make the script executable:

   sudo chmod +x /usr/local/bin/check_nginx.sh
   

3. Add a cron job to run it every 5 minutes (adjust the interval as needed):
   Run sudo crontab -e and add this line:

   */5 * * * * /usr/local/bin/check_nginx.sh
   

Method 3: Custom monitoring script (for more control) #

For advanced scenarios, you can write a more robust script that checks both the process and whether Nginx is responding to requests. For example:

#!/bin/bash
# Check if Nginx is listening on port 80/443
if ! curl -s http://localhost > /dev/null; then
    sudo systemctl restart nginx
    echo "$(date): Nginx was unresponsive, restarted it" >> /var/log/nginx_restart.log
fi

This ensures Nginx isn't just running but actually serving traffic.

内容的提问来源于stack exchange，提问作者yarek