Is there any reason why a network stream would fail before the timeout? I was using streams on an sbRIO-9632 and all throughout testing I had no problems. But, when I hooked the network up to NASA's dedicated network cable (which goes through one or two switches to eventually connect to my router, the same one used for testing; at NASA's Lunabotics Mining Competition; I think they said it was a virtual network), the streams, as far as I could tell, were failing between immediately and the 15 second timeout. If I recall correctly, it rarely made it beyond the timeout value.
Unfortunately, I did not have the time to successfully debug the system while connected to that network cable and switches so I was not able to get the exact error that was occuring. I do know that the error was originating from the sbRIO because on my computer it was showing that the network streams were closed.
If you are curious about the architecture being used, it was very similar to "Teleop - Host Acquisition" in the robotics module project wizard.
So I think the first thing to confirm here is that your computer can still see the sbRIO. What is the connection status to the sbRIO in MAX? Can you ping the RT target? Is the sbRIO on the same subnet as your computer?
Finally, have you been trying to look at the variables using the NI Distributed Systems Manager? Can you communicate with the card here?
Unfortunately, I've been unable to recreate the issue without being connected to the NASA setup (something I won't be able to do for the next year). The best that I can tell you is that the "fix" that I implemented was to simply re-establish the network stream and prevent it from stopping the rest of the program using a simple state machine for each end of each stream. It was able to re-connect fairly quickly so I never lost the connection for an appreciable amount of time (which would be required to see a loss in connection in MAX I'm assuming; maybe I'm wrong assuming this?).
Anyways, because I cannot replicate the issue and therefore test it again, I will try to not take up too much of your time on this one. I was just hoping to find out if there was any known circumstance that would cause the stream to fail/timeout prematurely.
It'll be pretty tricky to troubleshoot without actually being on the problematic network. There is no inherent limitation to using network streams with switches as far as I know, but you will need to ensure that each device is still configured correctly, i.e. that each device can successfully ping the other etc. The two devices must be on the same subnet as well.
Just a couple things to keep in mind if you run into this issue again.
Both devices communicating (sbRIO and laptop) were on the same subnet (192.168.0.xxx) and the ping was actually really good. IIRC, we did a constant stream of pings and all the times were very good and there might have been a timed-out ping every 100+ pings.
I guess I will just write the code as if it will always happen and maybe have some more logging for debugging it when we go down there next year.
Thanks for your time.