Compute Setup - trouble following tutorial

Hi guys, I have trouble following the Deployment to Production Servers Guide

As far as I can tell everything should be set up correctly
aws server has the rhino-compute service running
the …/version api gives me the current version of rhino and compute
core hour billing is active

However when I use the sample appserver to send requests, the only response is

{"message":"Internal Server Error","stack":"Error: Internal Server Error\n    at getParams (D:\\Code\\ComputeApp\\compute.rhino3d.appserver-main-opener\\src\\definitions.js:47:11)\n    at processTicksAndRejections (internal/process/task_queues.js:97:5)\n    at async D:\\Code\\ComputeApp\\compute.rhino3d.appserver-main-opener\\src\\routes\\template.js:46:14"}

everything works when i use a local compute server, but I’m not even sure where to start troubleshooting here
is there some compute logfile I can look at to figure this out?

Edit:
Is this maybe a duplicate of this:

So no local appserver + remote compute?
If yes: Why exactly is that? I thought the gh definitions are sent within the first request.

compute.geometry.exe logs to %tmp%\Compute\Logs

Yes, it sounds like the same issue. The appserver serves the Grasshopper file and sends a URL to this file when making requests to Compute. It was designed this way to reduce the amount of traffic - Compute only fetches the Grasshopper file once and caches it for future requests. If you’re running the appserver on localhost, then a Compute instance somewhere else will not be able to fetch the Grasshopper file.