Hi Remco,
NCrunch server is a new Windows 2019 Server machine (Version 1809 - OS Build 17763.107) without any other software installed. It's a dedicated server for NCrunch and both network and firewall seems ok.
I made some other stress-tests on the machine and I recorded a short video showing the situation when the service freeze:
https://www.screencast.com/t/KeSiZVoF
In the beginning, I can connect via telnet both from my machine and from the local server. After starting a huge number of tests from different machines and waiting for cca 1 hour, the NCrunch service closes the connections and I'm not able to reconnect to the machine for a while (you can see it cca at minute 3:10 of the video).
After that, I cannot connect via telnet anymore from my machine but it responds from localhost. Restarting NCrunch service from my VisualStudio does not work (Why? I suppose because NCruch service is not reachable from another machine).
I tried also to run the NCrunch console, but nothing changed :/
Checking the Windows event viewer I cannot find anything useful to understand what happens to NCruch service... So please, help us to understand why it freezes and what we can do or check to avoid this.
Maybe there is some special configuration for the NCrunch service to be restarted automatically or some special log we can enable in NCrunch to give you more clues on what happens... or something else we can do to solve this situation.
I had to restart manually the NCrunch service to make it work again, but this is not acceptable for us, we cannot do it 2-3 times all the days, I hope you understand.
BTW I tried also to stop the NCrunch service and run only NCrunch.GridNode.Console: it tells "Grid node successfully started and is listening for connections on port 41141" but if I try to check with 'netstat -an | find /i "listening"' it seems to not work, so what is the purpose of NCrunch Console?