Ah, makes sense! The server and frontend assume that everything is running at the “TLD” (the url) in your browser, ie. http://localhost:3000, so things get confused if they mismatch. I should’ve caught that from your first screenshots - they did show a separate vm in there 
Can you open the network logs in chrome and shoot a screenshot over once you’ve triggered the error? I don’t think there is a solution to this yet, though you can try to set the CANONICAL_URL env var to the ip of your linux box (192.168.1.11:3000).