Google Compute Engine实例重启后WordPress站点无法访问及apache2异常求助

Hey there, let's dig into why your GCE-hosted WordPress instance is inaccessible after a reboot—even though Apache is running. Here are the most likely culprits and step-by-step checks you can run:

Rebooting a GCE instance can mess with network settings that were working before:

• Temporary External IP Got Swapped: GCE uses temporary external IPs by default, which change every time you stop/start the instance. Head to the GCE Console to check your instance's current external IP, or run curl ifconfig.me on the instance to compare it with your domain's DNS record. If this is the issue, switch to a static external IP or update your domain's DNS settings.
• Firewall Rules Are Blocking Traffic: Double-check both GCE VPC firewalls and instance-level firewalls:
  • In the GCE Console, verify there's a firewall rule allowing HTTP (port 80) and HTTPS (port 443) traffic to your instance's tag or subnet.
  • On the instance, run sudo ufw status (if using UFW) or sudo iptables -L to confirm ports 80/443 are open to external traffic.

Just because Apache is "running" doesn't mean it's listening for external requests:

• Check Apache's Listening Status: Run sudo netstat -tulpn | grep apache2 to see if Apache is listening on 0.0.0.0:80/:::80 (allowing all IPs). If it only shows 127.0.0.1:80, it's only accepting local requests.
• Validate Apache Configs: Open your site's Apache config (e.g., /etc/apache2/sites-available/000-default.conf or your custom WordPress config) and make sure:
  • The Listen directive is set to 80 (or 443 for HTTPS)
  • The <VirtualHost> block uses *:80 (or your public IP) instead of localhost:80
• Check Apache Error Logs: Run sudo tail -n 20 /var/log/apache2/error.log to look for startup errors—like failed binding, missing config files, or permission issues accessing site files.

Reboots can sometimes reset file permissions, especially if Apache runs under a different user:

• Fix WordPress Directory Permissions: Ensure the WordPress root directory (usually /var/www/html) is owned by Apache's user (www-data):

  sudo chown -R www-data:www-data /var/www/html
  sudo chmod -R 755 /var/www/html
  sudo find /var/www/html -type f -exec chmod 644 {} \;
  

• Check Temporary Directory Permissions: Look for errors in Apache's error log related to accessing temporary directories (like /var/lib/apache2/run or PHP's temp folder) and fix permissions if needed.

WordPress can't load if it can't talk to its database, which might have issues after a reboot:

• Check Database Service Status: Run sudo systemctl status mysql (or mariadb if you use that). If it's not running, start it with sudo systemctl start mysql and enable auto-start on boot: sudo systemctl enable mysql.
• Verify Database Credentials: Open wp-config.php and confirm the DB_NAME, DB_USER, DB_PASSWORD, and DB_HOST match your database setup. Test the connection directly with:

  mysql -u your_wp_user -p
  

• Check Database User Permissions: Ensure your WordPress database user has full access to the WordPress database:

  GRANT ALL PRIVILEGES ON your_wp_db.* TO 'your_wp_user'@'localhost';
  FLUSH PRIVILEGES;
  

• Insufficient Resources: Rebooting might cause resource spikes. Run top or htop to check CPU/memory usage. If memory is maxed out, consider upgrading your instance size or enabling swap space.
• SELinux/AppArmor Blocks: If your instance uses these security modules, they might be preventing Apache from accessing files or network.
  • For SELinux: Run sudo sestatus—if it's in enforcing mode, temporarily switch to permissive with sudo setenforce 0 to test. If the site loads, adjust SELinux rules instead of disabling it entirely.
  • For AppArmor: Run sudo aa-status to check if Apache's profile is enforcing restrictions, and adjust the profile if needed.

内容的提问来源于stack exchange，提问作者Amal Antony