502 Error Troubleshooting Guide

This guide explains the most common causes of 502 Bad Gateway errors and how to fix them.

1. PHP-FPM Not Running​

A 502 Bad Gateway error is common when using Nginx. It means the web server (Nginx) did not receive a response from the backend (such as PHP-FPM, Node.js, or Python).

Steps to check PHP-FPM:

1. Verify that the PHP service is running. If Docker feature is enabled, go to Docker > Containers and check the status of the PHP container. If inactive, start it.

2. Restart Nginx​

Even if PHP-FPM is running, Nginx may have started before it and failed to connect on the first attempt, causing cached errors.

Fix: Restart Nginx.

• If using Docker feature: Go to Docker > Containers, disable Nginx, then enable it again.

• If you do not have access to Docker feature, simply add another domain in order to restart Nginx: Go to Domains > Add New, insert domain name and click on 'Add'.

3. Disable Varnish Cache​

Sometimes a 502 error is cached by Varnish, even if the backend issue is resolved. If you are using Varnish Cache:

Steps:

1. Go to Caching > Varnish and disable it for the affected domain.
2. Reload the website in your browser.

You can also bypass cache by adding parameters to the URL:

• Example: example.net?test or example.net?cachebypass

If disabling Varnish resolves the issue, stop and start it to clear all cached entries.

4. Verify VirtualHosts​

A misconfigured VirtualHost may cause Nginx to proxy requests to a backend that no longer exists. if you have access to the edit_vhost feature:

Steps:

1. Go to Domains > [Select Domain] > Edit VHosts.
2. Check the proxy target (e.g., PHP service name or container).
3. Ensure the service name is spelled correctly and still exists.

With these steps, you should be able to identify and fix the most common causes of 502 Bad Gateway errors.