LabView and -18001 error

[Eric L]

Hello,

I am a fairly new TestStand user, but have been using LabView for quite awhile. I have been running into a problem using TestStand and LabView together and have not been able to figure out.

My configuration is a Dell Optiplex GX270, with TestStand 3.0 and LabView 7.0. The PC is controlling a PXI chassis as well as several GPIB instruments.

The problem I am having is with intermittant -18001 errors when trying to run a sequence. Most often it will fail on the second or third time thru the sequence, but sometimes it will run for 10 or 20 loops. The problem is a -18001 error at a particular step, which calls a LabView vi that communicates with a power supply. Once the error occurs I need to exit both TestStand and Labview (which of
ten gets hung up) to recover.

I read "How to Fix LabVIEW -18001 Errors " but did not find anything there (doesn't seem to apply to intermittant errors). I also found some mention of TypeDef's being a potential problem (http://exchange.ni.com/servlet/ProcessRequest?RHIVEID=101&RNAME=ViewQuestion&HOID=50650000000800000019B20000&ECategory=LabVIEW.LabVIEW+General) but had no luck with that solution. I've also tried repairing the LabView installation, using the LabView runtime instead of the development version, etc. with no luck. The problem vi can be run hundreds of times in labview with no errors, so it appears to be related to using the LabView ActiveX server (I think).

What else is there to try?

[SachaE]

Hi,
could you please post your sequence (cut down to include the bare minimum for running the offending .vi), and the .vi and it's related files here so we can have a look into it?
What happens if you run that step interactively from the sequence (with and without process model)????

Thanks
Sacha Emery
National Instruments (UK)

[Eric L]

It turned out to be the VXIpnp driver for the Keithley 2400 sourcemeters. It was returning 5 elements into an array when I only requested 4, so it was overwriting memory.

It was hard to find because it wouldn't crash unless I ran two or more sequences in TestStand (was fine if I ran it in Labview by itself), plus the behavior kept changing, depending on how it ordered things in memeory (so if you changed something it might get better or worse, even though it was really unrelated).

The workaround was to increase the size of the results array for the extra dummy value, until a fixed version of the driver is available.

Eric

[Scott_Richardson]

Eric -
Thank you for posting what you found as your solution.

Sincerely,
Scott Richardson (NI)

[hegb_85]

Hi, I'm having the exact same error.

I was thinking about re installing labview but now i think that it is a programatic problem.

Do you recomend to search within the labview code to see if i have an error with the arrays or something like that?

I'll appreciate any ideas.

[hegb_85]

I include the screenshot i have and the *.txt file that windows generates.