I believe we've finally identified the root cause of the problem and the correct workaround. Please try the following:
1. Close TestStand.
2. In LabVIEW, go to Tools»Advanced»Clear Compiled Object Cache...
3. In the Clear Compiled Object Cache dialog, delete the LabVIEW target.
4. Open a new VI.
5. Drop the Simple Error Handler on the block diagram.
6. Run the VI.
7. Open TestStand and try your test again.
Please reply to this post with your results regardless of the outcome. Hope this helps!
I ran the affected test in isolation (using the run-time environment) and it worked. However, when I attempted to run the entire sequence a different step failed this time. Any ideas why?
I'm not sure I understand, so let's break this down:
First, are you saying that you followed my steps, then ran the specific error-prone step by itself, and it executed successfully? Just to confirm, did you follow my steps? If so, did they make a difference?
Second, a different step is failing. Is it throwing the same error? Are you running your test using the TestStand Sequence Editor or a custom TestStand User Interface? Does this new failing step call any of the error handler VIs (Simple Error Handler, General Error Handler)? If so, does commenting out or removing the error handler VIs resolve the error?
I followed your steps, they made a difference and the error-prone step passed. I then ran the full sequence of tests and a different test in the sequence failed. This step throws up the exact same error as the previously failing one did.
I'm running the tests in TestStand (for development) and in a custom user interface but the error is the same in both environments once the run-time engine is selected.
This new failing VI has 100's of VI's in it but I had a quick look and it does use the General Error Handler CORE.vi. I can't comment out this error handler as the actually module code is not mine - I'm just writing the test sequence to run the tests.
Thanks for you help so far,
Could you provide a screenshot of the Clear Compiled Object Cache dialog from your LabVIEW 2010 installation? Additionally, please provide a screenshot of the following directory: My Documents\LabVIEW Data\VIObjCache\10.0
See screenshot below (I ran this cleanup as you requested earlier):
Also, the directory you mentioned goes a bit deeper:
C:\Users\optest\Documents\LabVIEW Data\VIObjCache\10.0\LabVIEW\i386 - this directory contains the following:
C:\Users\optest\Documents\LabVIEW Data\VIObjCache\10.0\LabVIEW\i386\FC\E297999342396E6F185AA22D7DC103 - this directory contains the following:
C:\Users\optest\Documents\LabVIEW Data\VIObjCache\10.0\LabVIEW\i386\FC\E297999342396E6F185AA22D7DC103\9E7ACB34BC3F81D01A96855D9CBECAE1 - this contains the following:
If the path contained in the sourcePath.txt file located in C:\Users\optest\Documents\LabVIEW Data\VIObjCache\10.0\LabVIEW\i386 matches the path of your LabVIEW 2010 directory, then you are now experiencing a new problem. Have you mass-compiled the new problematic VI? What happens if you open that VI in LabVIEW? Is there a "dirty dot" (*)?
The path in the sourcePath.txt is 'C:\Program Files\National Instruments\LabVIEW 2010'.
I have mass compiled the llb I'm using and the original test fails. If I then clear the cache in Labview the new test fails.
I may have to run all of my tests in developer mode for the time being but hopefully we can get to the bottom of this.
P.S. there is no dirty dot when I mass compile the relevant VI's.
I think we should probably look into this issue closer offline. Would you mind if I contacted you directly? You can private message me your contact information.
We got the same issue,and it got fixed by removing all other dependent files like,Error handler.vi etc and only top level vi is only loaded.This issue is faced only with TS 4.2.1,earlier when we were using TS 3.5 the issue was not seen.