Hello. Here is my situation: I have a remote Unix server (let's call it "server") which is not visible from outside its local network, to which I connect to from my home windows computer ("PC") via a login server ("login"). login is also Unix based.
So, for example, in PC I open the "Bash on Ubuntu on Windows" app, then connect via ssh to login and then connect via ssh to server. Or I could do the same with PUTTY. Both connections (PC to login and login to server) need my username and password for the remote network.
The thing is I've been trying to run Spyder locally on PC but connected to a remote kernel running on server, but I could not achieve this connection. I have followed many tutorials like this, this and this but I could not find a solution to effectively establish the correct tunnel between PC and server. These and other tutorials I've found do not account for the fact that server is not visible from the Internet and/or that I need to authenticate twice. Or I least I could not understand how to implement it in my case.
Spyder includes an automatic remote connection via ssh but only tunnels directly to the server, it does not account for jump hosts. I think the right way to do it would be by manually forwarding the ports via a PUTTY connection, but I tried different configurations and none of them worked.
Any help would be appreciated.
PS: Spyder in PC is running on Windows. I only use the "Bash on Ubuntu on Windows" app for console access.