I'm attempting to offer an AT/Modem-like interface around some hardware. Follwing this post I've the server establishing a pty using openpty().

Now I'm able to talk to the server not surprisingly having a client application that open the slave and conveys via read() and write() calls.

However would like to have the ability to use either the screen command or minicom to problem instructions manually towards the slave. Nevertheless the server never appears to get data when attempting to get this done. Can there be something significant I'm missing with this particular approach?

Therefore the client application works together with the hardware, but minicom doesn't. Most likely the customer application is correctly setting the baud rate and quantity of stop bits (a minimum of individuals), and also you haven't requested minicom to complete exactly the same. You need individuals configurations right.