Ambiguous resource closing and examples do not close after completion. about node-ethernet-ip HOT 1 OPEN

If this is a connected session, then the PLC will terminate the connection on its end after some period of inactivity. In my own library, this is set to about 5 seconds in the set up for ForwardOpen. I have done tests and that timeout is pretty accurate. If you try to use the PLC TCP connection again after that, you will not be able to.

If this is an unconnected (UCMM) session, then I am not sure what will drive the timeout on the PLC side, but there will be one.

I ended up just terminating the connection internally in my library after about 5 seconds of inactivity and closing everything down. If a tag is read again, I reopen the connection transparently (somewhat transparently, as the delay is seen by the client).

This behavior of transparent reconnection solved a lot of weird complaints people had. So I suspect that the PLC disconnecting was a much bigger problem than I saw in my own testing. I have since added several tests that intentionally wait 30 seconds or more between tag requests.

I hope this helps!

from node-ethernet-ip.