Ping :)

Also got into this issue today. Also happens with slow log.

This is due how fpm_main.c replaces request_uri with SCRIPT_NAME.

Any idea how nginx/fpm users can work around it?

It happens with basically such setup (notice internal part).

    location / {
        try_files $uri /index.php$is_args$args;
    }

    location ~ ^/index\.php(/|$) {		
        fastcgi_pass php74;
        fastcgi_split_path_info ^(.+\.php)(/.*)$;
        include fastcgi_params;
        fastcgi_param SCRIPT_FILENAME $realpath_root$fastcgi_script_name;
        fastcgi_param DOCUMENT_ROOT $realpath_root;
        internal;
    }

And if you look at status page you will notice that example.com https://example.com/login
renders as:
request URI:          /index.php

Even though in php I can see proper REQUEST_URI variable inside $_SERVER vars.

A workaround is to define the FastCGI variable SCRIPT_NAME: in nginx:
 fastcgi_param  SCRIPT_NAME $request_uri;
As described in the blog post jolicode.com/blog/how-to-see-full-request-uri-in-fpm-status

I'm not sure of the exact meaning of SCRIPT_NAME (versus REQUEST_URI), but perhaps the FPM status page could directly use REQUEST_URI in the column request_uri, or possibly use REQUEST_URI and, if empty, have a fallback to SCRIPT_NAME?

Another intermediary solution would be to add both in the JSON/XML/etc result, so that it is easier to display according to its own configuration, by modifying this file https://github.com/php/php-src/blob/master/sapi/fpm/status.html.in .