small issues after upgrading to debian 13 (app-vhost+php)

The update itself went smoothly; manual intervention was only required when switching from PHP 8.2 to 8.4. Everything also worked fine with Dovecot.
But, there are some small issues i want to check:
I’m having a problem accessing the app vhost — both Webmail and phpMyAdmin are throwing a 502 error.

The cause appears to be a conflicting configuration between Nginx and PHP, or possibly something missing on my system. The migration was done at the time from an existing Debian 11 + ISPConfig + Apache setup to a freshly installed Debian 12 using the ISPConfig Migration Toolkit.

In the Nginx config, PHP is configured for the app vhost as follows:

Code:

server {
        listen 8081 ssl http2;
        listen [::]:8081 ssl http2 ipv6only=on;
..........
        location ~ \.php$ {
..........
               # PHP only, required if PHP was built with --enable-force-cgi-redirect
               fastcgi_param   REDIRECT_STATUS         200;
               fastcgi_pass unix:/var/lib/php8.4-fpm/apps.sock;
               fastcgi_index index.php;
..........

However, in the locations for webmail/squirrelmail/phpMyAdmin, the configuration looks like this — the socket is inactive and access via TCP is enabled instead:

Code:

fastcgi_pass 127.0.0.1:9000;
#fastcgi_pass unix:/var/lib/php8.4-fpm/apps.sock;

PHP 8.4 is running entirely in socket mode on the system.

Another noticeable thing in the FPM pool is a placeholder variable:

Code:

[apps-{fpm_domain}]
listen = /var/lib/php8.4-fpm/apps.sock

I’ve already rewritten the entire configuration via resync, but the behavior remains the same. Now I’m unsure whether there is an error in the system setup itself or whether this is an issue with the ISPConfig configuration. Manually switching to the socket causes the apps to work — what would be the correct approach to make these changes persistent?

 

Ok thank you!