Remotely Stream ended ISP/NGINX

[CFox243]

So, iv been trying to track this down for a while now. Mediacom connection ends the stream every 30-60 seconds, but no other ISP does. Seems to be Mediacom on a wireless network that causes it, but the system connection never drops out so not sure.
not sure if it has something to do with the area or if I have my Nginx proxy miss configured. local works fine, even going through the proxy.
but if someone is on Mediacom they disconnect constantly. AT&T connection seems to be fine. any help would be greatly appreciated.

Proxy advanced settings

  # HSTS (ngx_http_headers_module is required) (63072000 seconds = 2 years)
  add_header Strict-Transport-Security "max-age=63072000; preload" always;

    proxy_set_header Upgrade $http_upgrade;
    proxy_set_header Connection $http_connection;
    proxy_http_version 1.1;

    # Proxy!
    include conf.d/include/proxy.conf;
  proxy_send_timeout 300;
  proxy_read_timeout 300;
  }

location /_blazor {
   proxy_set_header Connection $http_connection;
    proxy_http_version 1.1;

    # Proxy!
    include conf.d/include/proxy.conf;

  proxy_send_timeout 300;
  proxy_read_timeout 300;
  }

location /AgentHub {
  proxy_set_header Connection $http_connection;
    proxy_http_version 1.1;

    # Proxy!
    include conf.d/include/proxy.conf;

  proxy_send_timeout 300;
  proxy_read_timeout 300;
  }

location /ViewerHub {
 proxy_set_header Connection $http_connection;
    proxy_http_version 1.1;

    # Proxy!
    include conf.d/include/proxy.conf;

  proxy_send_timeout 300;
  proxy_read_timeout 300;
  }

location /CasterHub {
 proxy_set_header Connection $http_connection;
    proxy_http_version 1.1;

    # Proxy!
    include conf.d/include/proxy.conf;

  proxy_send_timeout 300;
  proxy_read_timeout 300;
  }

The host connection is 1gb up/down

[CFox243]

Based on wireshark the website seems to reload every 30 seconds.
when i go to a PC and try to change device group when i select a different group but don't hit save and let it sit within 30 seconds it will auto refresh back to what it was.

Remotely.mov

[CFox243]

It seems like the Mediacom Modem/router might be closing a port that is needed after connection.
The issue is that it is an Employee's Personal network, so I can't really make changes to it.
I'm having a hard time reproducing the issue in a dev environment.
I had them use a hotspot on their phone, and it connected without issue.
Their home internet and connection are plenty fast for the job.

[CFox243]

For people looking for NGINX Advanced Variables

   location / {
        proxy_pass         http://ip:5000;
        proxy_http_version 1.1;
        proxy_set_header   Upgrade $http_upgrade;
        proxy_set_header   Connection keep-alive;
        proxy_set_header   Host $host;
        proxy_cache_bypass $http_upgrade;
        proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header   X-Forwarded-Proto $scheme;
        proxy_send_timeout 30m;
        proxy_connect_timeout 3600s;
        proxy_read_timeout 30m;
        client_max_body_size 10M;
    }
    location /_blazor {	
        proxy_pass http://ip:5000;
        proxy_http_version 1.1;	
        proxy_set_header Upgrade $http_upgrade;	
        proxy_set_header Connection "upgrade";	
        proxy_set_header Host $host;	
        proxy_cache_bypass $http_upgrade;	
        proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;	
        proxy_set_header   X-Forwarded-Proto $scheme;	
        proxy_send_timeout 30m;
        proxy_connect_timeout 3600s;
        proxy_read_timeout 30m;
        client_max_body_size 10M;
    }	
    location /AgentHub {	
        proxy_pass   http://ip:5000;
        proxy_http_version 1.1;	
        proxy_set_header Upgrade $http_upgrade;	
        proxy_set_header Connection "upgrade";	
        proxy_set_header Host $host;	
        proxy_cache_bypass $http_upgrade;	
        proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;	
        proxy_set_header   X-Forwarded-Proto $scheme;	
        proxy_send_timeout 30m;
        proxy_connect_timeout 3600s;
        proxy_read_timeout 30m;
        client_max_body_size 10M;
    }	
    location /ViewerHub {	
        proxy_pass   http://ip:5000;
        proxy_http_version 1.1;	
        proxy_set_header Upgrade $http_upgrade;	
        proxy_set_header Connection "upgrade";	
        proxy_set_header Host $host;	
        proxy_cache_bypass $http_upgrade;	
        proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;	
        proxy_set_header   X-Forwarded-Proto $scheme;	
        proxy_send_timeout 30m;
        proxy_connect_timeout 3600s;
        proxy_read_timeout 30m;
        client_max_body_size 10M;
    }	
    location /CasterHub {	
        proxy_pass   http://ip:5000;
        proxy_http_version 1.1;	
        proxy_set_header Upgrade $http_upgrade;	
        proxy_set_header Connection "upgrade";	
        proxy_set_header Host $host;	
        proxy_cache_bypass $http_upgrade;	
        proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;	
        proxy_set_header   X-Forwarded-Proto $scheme;	
        proxy_send_timeout 30m;
        proxy_connect_timeout 3600s;
        proxy_read_timeout 30m;
        client_max_body_size 10M;
    }

[akorb90]

I had the same issue with the following "basic" config:

# Upgrade WebSocket if requested, otherwise use keepalive
map $http_upgrade $connection_upgrade_keepalive {
    default upgrade;
    ''      '';
}

################## upstream config ####################

upstream remote_public_domain {
    server server:5000;
    # Improve performance by keeping some connections alive.
    keepalive 10;
}

################### server config #####################

server {
    # HTTP server config
    listen 80;
    server_name remote.public.domain;

    # 301 redirect to HTTPS
    return 301 https://$host$request_uri;
}

server {
    listen                  443 ssl http2;
    server_name             remote.public.domain;

    client_max_body_size    50M;

    access_log              /var/log/nginx/access_remote.log combined;
    error_log               /var/log/nginx/error_remote.log warn;

    ssl_certificate         /root/.acme.sh/public.domain/fullchain.cer;
    ssl_certificate_key     /root/.acme.sh/public.domain/public.domain.key;
    ssl_trusted_certificate /root/.acme.sh/public.domain/ca.cer;

     # Specify SSL protocols
    ssl_protocols TLSv1.2 TLSv1.3;

    # Specify SSL ciphers
    ssl_ciphers HIGH:!aNULL:!MD5;
    ssl_prefer_server_ciphers on;

    # Enable HSTS
    add_header Strict-Transport-Security "max-age=31536000; includeSubDomains" always;

    # Optimize SSL settings
    ssl_session_cache shared:SSL:10m;
    ssl_session_timeout 10m;


    location / {
        include                 conf.d/proxy_set_header.inc;
        proxy_pass              http://remote_public_domain;
    }
}

Switching to your proxy config immediatelly fixed the issue.

You should think about providing it in the docs as a reference for an nginx conf.

Router was an OpenWRT device though, traffic coming in through cloudflare proxying.

[clincher61]

I had Remotely working with a basic config in Ngnix Proxy Manager. Everything was working fine except all clients look like they're coming from the proxy IP address as their Public IP. After switching to Cfox243's advanced block, I'm still getting the same Public IP as my proxy. I've added my proxy to the Known Proxy as well. Hmm?

[clincher61]

Scratch that! A restart of Remotely server fixed all of that. Cfox243's advanced NGINX block should be added to the docs!