If someone with some more SSH knowledge and experience can get back with me, I’d appreciate it. I am working with a customer and he is having problems with my software connecting via SSH. I’m using an SSH class that Kem Tekinay helped me develop a couple years ago. I’m connecting properly to their device. However, it’s not letting me authenticate properly and is just hanging. The SSH class uses an interactive shell for Mac and Linux (we use plink.exe for Windows). The platform my customer is running is Linux. Here’s part of the output from their console:
debug~: SSH2_MSG_NEWKEYS sent debug~: expecting SSH2_MSG_NEWKEYS debug~: SSH2_MSG_NEWKEYS r e c e i v e d debug~: SSH2_MSG_SERVICE_REQUEST sent debug~: SSH2 MSG_SERVICE_ACCEPT r e c e i v e d debug~: Authentications that can continue: keyboard-interactive,password debug~: Next authentication method: keyboard-interactive Password:
We are trying to send the login information automatically. This is all supposed to take place behind the scenes. We send the user name and password and then commands to manipulate their device. Works fine when connecting via Telnet and I have other customers using SSH with no problem. But their configuration appears different.
What I think I am reading is that the SSH server is expecting actual keyboard input before allowing the login to proceed (i.e.: next authentication method: keyboard-interactive). So since we are sending things automatically, the keyboard is not being used and the server is therefore ignoring any input. Do I have that understanding correct?
How can I change this? Is this something they need to change on the host end? Or can I do something in the launching of the SSH client in the shell? Or is there a way for Xojo to send keystrokes?
Any help is great.