误删tmp目录nginx.pid及tmp文件内容，如何重建并解决Nginx报错？

Hey there, let's work through these two Nginx issues step by step—they’re both straightforward to fix!

First up, the error about failing to open /var/log/nginx/error.log and access.log likely happened because those files (or maybe the entire /var/log/nginx directory) were accidentally removed during your cleanup. Here’s how to fix it:

• First, make sure the /var/log/nginx directory exists. If it was deleted, create it with:

  sudo mkdir -p /var/log/nginx
  

• Next, create the missing log files:

  sudo touch /var/log/nginx/error.log /var/log/nginx/access.log
  

• Critical step: Set the correct ownership so the Nginx process can read/write these files. Nginx typically runs as the nginx or www-data user—check which one with ps aux | grep nginx, then run:

  # Replace nginx:nginx with www-data:www-data if your Nginx uses that user
  sudo chown -R nginx:nginx /var/log/nginx
  

• Now test your configuration again—this should resolve the log-related errors:

  sudo nginx -t -c /etc/nginx/nginx.conf
  

The nginx.pid file doesn’t need to be manually written—Nginx creates it automatically when it starts. Here’s what to do:

• First, confirm that the /tmp directory exists and has the right permissions (since you deleted its contents, the directory itself should still be there, but double-check):

  ls -ld /tmp
  

  The permissions should show drwxrwxrwt—if not, fix them with:

  sudo chmod 1777 /tmp
  

• Next, check where your Nginx configuration expects the pid file to be. Run this grep command to verify:

  grep -i pid /etc/nginx/nginx.conf
  

  If it shows pid /tmp/nginx.pid;, that’s the path we need. If it’s set to another location (like the default /var/run/nginx.pid), adjust the verification step below accordingly.
• Finally, restart Nginx. This will trigger Nginx to generate the pid file in the configured path:

  # For systemd systems (most modern distros like Ubuntu 16.04+, CentOS 7+)
  sudo systemctl restart nginx
  # For older systems using init.d
  sudo service nginx restart
  

• Verify it worked by checking if the pid file exists:

  ls /tmp/nginx.pid
  

A quick heads-up: Avoid deleting all files in /tmp blindly—many services rely on temporary files, sockets, and pid files stored there. If you need to clean up /tmp, use tools like sudo tmpwatch that safely remove old unused files instead of wiping everything.

内容的提问来源于stack exchange，提问作者enigmq