After a full check of all hosted sites, we confirm that everything is back to normal. All services are operating correctly.

This incident followed an automatic update of certain PHP versions and the integration of an exception by CloudLinux. During a routine reload of the PHP-FPM services in the morning, several versions ended up listening on the same port, which prevented some of them, particularly the older ones, from starting. This is why only some sites were affected.

Resolving it proved especially delicate: the fix had to work in tandem with Plesk, which required careful work to restore access without affecting the rest of the platform. Our teams put a reliable workaround in place, pending the definitive fix released by the vendor.

We apologize for the inconvenience and thank you for your patience throughout its resolution.

We have identified a bug introduced by CloudLinux in certain PHP versions, which explains why the issue only affected some sites. A first fix was deployed at around 09:10\. We are no longer seeing any errors at this stage. Our teams are continuing a detailed check of all hosted sites to make sure nothing is overlooked.

We are currently experiencing an incident affecting access to some websites hosted on our FRWEB14 server. Our technical teams are fully mobilized and are actively working to restore service as quickly as possible.