How to fix 500 Internal Server Error on Apache

A 500 Internal Server Error tells the visitor that something went wrong, but not what. The real message is always in a log. This guide shows where to look, how to recognise the common causes on an Apache server, and how to isolate the problem when the logs are not clear.

Before you start

• A Linux VPS running Apache 2.4 on Ubuntu, Debian, AlmaLinux or Rocky Linux.
• SSH access with sudo.
• A recent backup of the site files, or at least a copy of any file you plan to edit.

Step 1: Read the error log

Reload the failing page, then read the last lines of the error log at once, while the error is fresh:

sudo tail -n 30 /var/log/apache2/error.log   # Ubuntu/Debian
sudo tail -n 30 /var/log/httpd/error_log     # AlmaLinux/Rocky

If the site has its own ErrorLog in the virtual host, read that file. If you run PHP-FPM, also check the FPM log, for example /var/log/php8.3-fpm.log on Ubuntu or /var/log/php-fpm/www-error.log on AlmaLinux.

The sections below match the most common messages.

Bad.htaccess directives

This is the most frequent cause. The log shows something like:

[core:alert] /var/www/example.com/public/.htaccess: Invalid command 'RewriteEngine', perhaps misspelled or defined by a module not included in the server configuration

The line names the file and the directive. Two common versions:

• Invalid command 'RewriteEngine': mod_rewrite is not enabled. On Ubuntu and Debian, run sudo a2enmod rewrite and restart Apache. On AlmaLinux and Rocky it is loaded by default; check /etc/httpd/conf.modules.d/00-base.conf.
• Invalid command 'php_value': php_value and php_flag only work with mod_php. Under PHP-FPM, remove those lines and put the settings in php.ini or a .user.ini file in the web root.

To prove .htaccess is the cause, rename it and reload the page:

sudo mv /var/www/example.com/public/.htaccess /var/www/example.com/public/.htaccess.bak

If the page loads (perhaps with broken links), restore the file and remove or fix the bad line.

Missing Apache modules

Directives such as Header, ExpiresActive or ProxyPass need their module. List what is loaded:

sudo apachectl -M | sort

Enable a missing module on Ubuntu or Debian:

sudo a2enmod headers expires
sudo systemctl restart apache2

On AlmaLinux and Rocky, most modules are loaded from /etc/httpd/conf.modules.d/. Some, such as mod_ssl, come as separate packages installed with dnf.

PHP fatal errors

When PHP crashes, Apache often returns a 500 with a blank page. The log shows a line such as:

PHP Fatal error:  Uncaught Error: Call to undefined function mysqli_connect() in /var/www/example.com/public/db.php:4

Common causes and fixes:

• Undefined function: a PHP extension is missing. Install it, for example sudo apt install php8.3-mysql or sudo dnf install php-mysqlnd, then restart PHP-FPM or Apache.
• Allowed memory size exhausted: raise memory_limit in php.ini, or find the plugin that uses too much memory.
• Syntax error after an edit: check the file with php -l /path/to/file.php.

If you see nothing in the logs, make sure PHP logs errors. In php.ini set log_errors = On and keep display_errors = Off on a live site.

Permissions and ownership

Scripts that Apache cannot read, or a CGI script without execute permission, can cause a 500. On PHP-FPM, a socket that Apache cannot open produces errors such as AH01079: failed to make connection to backend. Check:

ls -l /run/php/
namei -l /var/www/example.com/public/index.php

Files should normally be 644 and directories 755. Make sure the FPM pool's listen.owner and listen.group let Apache use the socket.

On AlmaLinux and Rocky, also check SELinux with sudo ausearch -m avc -ts recent. A denial there means a file label or boolean needs fixing, not a permission change.

Isolate the cause

When the log is unclear, narrow it down in this order:

1. Serve a static file. Create test.html in the web root. If it loads, Apache itself is fine and the problem is PHP or the app.
2. Serve a minimal PHP file:

   echo "<?php echo 'ok';" | sudo tee /var/www/example.com/public/ok.php

If ok.php works but the site does not, the problem is in the application code or its config.

1. Disable plugins or modules in the app one at a time. For WordPress, rename the wp-content/plugins folder.
2. Remove the test files when you finish.

Check it works

1. Reload the page. It should return content, not the 500 page.
2. Confirm the status from the command line: curl -I https://example.com/ should show 200.
3. Tail the error log while you click around the site. No new errors should appear.

Troubleshooting

• Nothing in the Apache error log: the request may be handled by a different vhost with its own log, or by PHP-FPM. Run sudo apachectl -S to find the right vhost, and read the FPM log.
• 500 only after a plugin or theme update: the update introduced a PHP error. Roll it back or disable it, then check the PHP version it needs.
• 500 appears at random under load: PHP-FPM may be running out of workers or memory. Look for max_children warnings in the FPM log.
• Request exceeded the limit of 10 internal redirects: a rewrite rule loops. Fix the rule, often by adding RewriteCond %{REQUEST_FILENAME} !-f.

Next steps

• Learn your logs in depth: Find and read Apache access and error logs.
• Make .htaccess work correctly: Enable.htaccess and mod_rewrite in Apache.
• WordPress-specific causes: Fix the WordPress critical error.
• If you need help with your server, open a ticket.