Dutch,
I'm glad to hear you got your project working, however you did it.
For anyone else following this thread with connection issues:
A friend of mine with much more VM experience than I described how he got this working in his network and I was, mostly, able to recreate his success.? It turns out that if you get the CTCE defined properly and attached in Hercules before either VM is IPLed, when they are IPLed this will work.? His theory is that the something in the IPL process initializes the CTC in a way that allows this to be successful.
As it turns out, I also had a configuration issue in my CTCE set up that seems incompatible with this connection type, even though it works with VTAM.
In my systems, I had the CTCEs defined as follows in hercules.cnf.? While this works fine for VTAM CTC links, this does not work for RSCS NJE links.
VM/SP - 600 CTCE 30802 602=Pi2 30804 ATTNDELAY 200 # link to JS04/602
VM/ESA - 602 CTCE 30804 600=Pi2 30802????? # link to JS02/600
Notes: 600 and 602 are the unit addresses on each VM, 30802 and 30804 are port numbers for the TCPIP connection and Pi2 is the hostname for the Hercules host computer.
My friend's configuration use the following format, which also worked for me when I tested it.
VM/SP - 500 ctce 5000 192.168.1.2 6000
VM/ESA - 600 ctce 6000 192.168.1.2 5000
Notes: 500 and 600 are the unit addresses, 5000 and 6000 are TCPIP port numbers and 192.168.1.2 is the IP address of the Hercules host computer.
If you include the device definitions in your Hercules configuration file, that will guarantee they are in place before the guest VM is IPLed.? In my testing, I brought both Hercules instances up and had the CTCs attached and the connection up between them before I IPLed either guest.? Then I IPLed, brought up RSCS, defined the links and when I started them, everything worked.
Jeff
