Getting VI_ERROR_IO error on serial port from VISA 3.0

[HCL]

I am using COM1 (via the VISA driver) to communicate and test my UUT. In the middle of the test I would send a "reboot" command to reboot the UUT and keep reading the port until the promt shows up. This test runs fine when I have VISA 2.5/2.6 installed on the computer. Recently, I have upgraded from LabVIEW6.1 to LabVIEW7.0. The VISA driver is also upgraded to 3.0 during the installation. However, the test would cause a VI_ERROR_IO error after the UUT is rebooted. This forced me to fall back to older drivers to resolve the problem. So why is the VISA 3.0 causing this I/O error during the UUT reboot? Is there any driver behavior change in this area? Can I get a fix for this?

[A.S.]

Hi,
What would be helpful is to run NI-SPY to get a capture of all the communication that is occurring in the ports. If you can do a SPY capture that displays where you get that error, and post it here, that'd be very helpful in narrowing down the problem.

[David@Work]

I've been struggling with the same thing. Contacted NI, and was never able to get a resolution. I have kept VISA 3.01, but unwired error clusters from serial port reads. (Quite dangerous I know). NI SPY comparisons between machines running 2.5, and 3.01 showed nothing different. I'm sure glad to hear I'm not the only one having trouble.

[Shannon R]

HCL,

As A.S. states, it would be helpful to have the NI Spy Capture of the actual commands that are being sent in your case. Also, what do you mean by the "UUT is rebooted"? Does this mean after one unit is done testing, or is there an actual code module that reboots your UUT?

If this is a driver related problem, then you should be able to reproduce this in any programming language. Have you seen similar behavior in LabVIEW without using TestStand? Does it only happen with your sequences?

Thanks,
Shannon R.
Applications Engineer
National Instruments