Randomly getting NGINX 502 Bad Gateway when connecting to ERPNEXT from mobile devices

Question

does anyone have experience with ERPNext in combination with an NGINX server in front of it? I am getting 502 Bad Gateway messages nearly every time i am accessing ERPNext from mobile devices via internet.

My setup

• Multiple sub-domains directed to the public ip-address of my firewall -> 443 packets forwarded to my NGINX server -> NGINX server dealing with the different domains and routing the requests into my local network
• Except for the erpnext proxy_pass all other domains are routed without any issues, not matter if i access from PCs or mobile devices
• NGINX 1.18.0 running on a Ubuntu 20.04.6 LTS
• ERPNext running on a Ubuntu 22.04.4 LTS

What i have troubleshooted so far:

• I can force the 502 Bad Gateway pretty easily by connecting to erpnext from a mobile device, do some stuff within the app and then switch networks from WiFi to 5G and back. Atfer that i immediately get the 502 Bad Gateway.
• Mobile networks are mostly working with IPv6, WiFi in my case 99% IPv4
• When accessing erpnext from local network i am not getting any issues as my dns is directing any client directly to the erpnext host and bypasses the nginx server
• The NGINX error.log shows following message:

2024/03/02 11:06:00 [error] 631#631: *28 connect() failed (111: Connection refused) while connecting to upstream, client: 109.42.xx.xx, server: erp.xx.xx, request: "GET /login HTTP/1.1", upstream: "https://192.168.xxx.xxx:443/", host: "erp.xx.xx", referrer: "https://erp.xx.xx/app/home>

My Nginx config:

http {

        ##
        # Basic Settings
        ##

        more_set_headers 'Server: ';

        sendfile on;
        tcp_nopush on;
        tcp_nodelay on;
        keepalive_timeout 65;
        types_hash_max_size 2048;
        server_tokens off;

        # server_names_hash_bucket_size 64;
        # server_name_in_redirect off;

        include /etc/nginx/mime.types;
        default_type application/octet-stream;

        ##buffer policy
        client_body_buffer_size 1K;
        client_header_buffer_size 1k;
        client_max_body_size 1k;
        large_client_header_buffers 2 1k;
        ##end buffer policy



        ##
        # SSL Settings
        ##

        ssl_protocols TLSv1.2 TLSv1.3; # Dropping SSLv3, ref: POODLE
        ssl_prefer_server_ciphers on;
        ssl_ciphers "EECDH+ECDSA+AESGCM EECDH+aRSA+AESGCM EECDH+ECDSA+SHA384 EECDH+ECDSA+SHA256 EECDH+aRSA+SHA384 EECDH+aRSA+SHA256 EECDH+>

        ##
        # Logging Settings
        ##

        access_log /var/log/nginx/access.log;
        error_log /var/log/nginx/error.log;

        ##
        # Gzip Settings
        ##

        gzip on;

        # gzip_vary on;
        # gzip_proxied any;
        # gzip_comp_level 6;
        # gzip_buffers 16 8k;
        # gzip_http_version 1.1;
        # gzip_types text/plain text/css application/json application/javascript text/xml application/xml application/xml+rss text/javascr>

        ##
        # Virtual Host Configs
        ##

Site-enabled:

server {
        listen 443 ssl; # managed by Certbot
        listen [::]:443 ssl;
        ssl_certificate /etc/letsencrypt/live/erp.xx.xx/fullchain.pem; # managed by Certbot
        ssl_certificate_key /etc/letsencrypt/live/erp.xx.xx/privkey.pem; # managed by Certbot
        include /etc/letsencrypt/options-ssl-nginx.conf; # managed by Certbot
        ssl_dhparam /etc/letsencrypt/ssl-dhparams.pem; # managed by Certbot


        server_tokens off;

        server_name erp.xx.xx;
        location / {
                proxy_pass      https://192.168.xxx.xxx/;

                proxy_set_header Host $host;
                proxy_set_header X-Real-IP $remote_addr;
                proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        }
        client_max_body_size 0;

}

I hope somebody can help me with this.. Thank you!