From Friday, April 19th (11:00 PM CDT) through Saturday, April 20th (2:00 PM CDT), 2024, ni.com will undergo system upgrades that may result in temporary service interruption.

We appreciate your patience as we improve our online experience.

NI TestStand

cancel
Showing results for 
Search instead for 
Did you mean: 

TestStand has encountered a problem and needs to close...

Hi everyone,
 
I have an issue with TestStand crashing and would appreciate some guidance from some wise TS gurus...
 
I am using Win XP SP2 with TS3.5 Installed (along with everything) from the February issue of a Professional Developer Suite.
 
I have a few large sequence files that are used to production test a family of products.  After a couple of executions of these sequences TestStand crashes, returning the error dialog seen in the attached file.  The problem was first seen on a PC with pretty much the same setup as above, but I have also seen it on a PC built just for deployment running the sequences in a custom (and an NI shipped) Operator Interface.  Both these PCs were, however quite old and very slow so I put the issues down to lack of system resources etc.  More recentally I have seen it on the brand new PC I have setup, which is when I really started worrying and experimenting.
 
The crash only occurs when I am running sequences for a given family of products.  These products/sequences utilise two dll's written (by our in-house software department) to communicate with these products.  The dll's are called using the ActiveX/COM adaptor within TS.  The first call to a dll creates a new object ref which is stored in a file global which is then used by subsequent calls to that dll.  I have made several experimental sequences and feel the source of the problem is these dll's (although im not 100% convinced?!).  I have created sequences calling other dll's (those included as examples / demos by NI) and have not seen the problem.  The crash can occur at almost any time - not necessarily during execution, but while editing the sequence.  The only prerequisite for the error is that one of the above metioned sequences has been executed alteast once.  When operating TS immediately after a reboot the problem takes longer to occur, but on subsequent restarts of TS the error happens sooner.
 
Is there anything you guys can recommend for me to try - to gain more information about what the cause of this error might really be?  If I decide it is these dlls's I need to supply our SW dept with some info other than "I think they crash TestStand".
 
Is there a known bug within TS that causes this kind of problem with some dll's?  And have any of you found a fix?
 
I am planning to try to use these dll's directly from a VB frontend (i.e. NOT using TS) that I am planning to write purely for experimentation - to see if I see the problem still rears its ugly head (May do the same in LabVIEW).  However these COM Classes are sold with the hardware for third parties to use, and to date no issues have been reported (I doubt any of them will be using TS).
 
I hope someone can help or point me in the right direction (any direction would be helpful!).  Any help would be much appreciated!
 
Regards,
Barry
0 Kudos
Message 1 of 4
(3,049 Views)
Original message wouldnt post with attachments - trying again....
0 Kudos
Message 2 of 4
(3,045 Views)
0 Kudos
Message 3 of 4
(3,042 Views)

Hi Bazza,

In my early days of TestStand, I saw these types of errors a number of times.  Everytime, I discovered that the source of the problem was either memory leak from a dll or a bad pointer within a Labview function call to a dll.  The memory leak was detected by the fact that TestStand could run the tests a number of times before displaying the error messages.   Whereas the pointer caused an immediate crash.

JLV

Message 4 of 4
(3,004 Views)