Elusive wiretbl.cpp error

[Nanopit]

I have written a very massive VI (the llb is just shy of 14 MB), and I've gotten a very elusive error twice in the last month and a half.  The first time, Labview just bombed shortly thereafter, but the second time, I got a log file upon restarting of the program (following a reboot).  The log file has several identical copies of the following code:

c:\builds\penguin\labview\branches\Jupiter\dev\source\diagram\wiretbl.cpp(3271) : DWarn: Attempt to use bad index for wire (3005260^+AB0A814). in C:\Documents and Settings\lockhp\Desktop\LTR_main_08072008.llb\LTR_main.vi
$Id: //labview/branches/Jupiter/dev/source/diagram/wiretbl.cpp#58 $
0x00A6B462 - LabVIEW <unknown> + 0
0x012D134E - LabVIEW <unknown> + 0
0x011C12AB - LabVIEW <unknown> + 0
0x00676955 - LabVIEW <unknown> + 0
0x00AC8E7F - LabVIEW <unknown> + 0
0x00AC9AA8 - LabVIEW <unknown> + 0
0x00ACB53E - LabVIEW <unknown> + 0
0x0049C139 - LabVIEW <unknown> + 0
0x0049A981 - LabVIEW <unknown> + 0
0x004AE4F0 - LabVIEW <unknown> + 0
0x7E418734 - USER32 <unknown> + 0
0x7E418816 - USER32 <unknown> + 0
0x7E4189CD - USER32 <unknown> + 0
0x7E4196C7 - USER32 <unknown> + 0
0x00468E64 - LabVIEW <unknown> + 0
0x0045D013 - LabVIEW <unknown> + 0
0x00505D85 - LabVIEW <unknown> + 0
0x00505FAB - LabVIEW <unknown> + 0
0x00505FF0 - LabVIEW <unknown> + 0
0x0149415F - LabVIEW <unknown> + 0
0x7C817067 - kernel32 <unknown> + 0
0x00000000 - LabVIEW <unknown> + 0

The symptoms surrounding the error are (bear in mind these may just be reprecussions of the error as this is an "always running" system and the errors have both happened overnight on the weekend and were found on Monday):

The Labjack U12 VIs I have employed have a popup window saying the device can't be found, but only for some of the Labjack units (I have 4 of them, all programmed identically with a device ID to distinguish each of them).  The Labjacks are accessed with a USB cable for each Labjack unit and referenced with a "refnum" approach.

A home-made VI to read the serial port of the computer to get data says that accessing the serial port is timing out.  The serial port is accessed using the VISA protocol.

The problem which I think is occurring is that the serial port access occurs directly after we access and read an Agilent 34970A DAQ unit using a home-made VI running on a GPIB protocol.  I have a suspicion that the transition between our DAQ and serial access VI's doesn't wait for the communication lines to clear before starting the next communication.  Thus, it's only a matter of time before a single slow moment on the computer causes this error (infrequently as it's only happened twice since early July).

In order to clear this error and allow the system to work correctly, I need to reboot the computer (otherwise the computer can't find the Labjacks, even if I unplug and replug the USB cables connecting/powering the Labjack units).

I've added a 2 second delay between accessing the Agilent DAQs and the serial port, and I've changed the home-made DAQ VI's for the ones from Agilent (which use the VISA protocol and have an initialize and close step for each DAQ access to make sure the comm lines are clear), but the time between errors is long enough that it's difficult to gauge if this is the fix we're looking for. 

This error message is odd ground for me, and I'm wondering if someone could please offer some help with understanding it to make sure I'm heading in the right direction for fixing the cause of the error.

[Nanopit]

Almost forgot to say that the computer specs are:

Labview 8.5

Core2 duo cpu

2 Gigs of RAM

I can get more specific info if it's needed.

This computer is dedicated to only running this program.

[jpaul]

Hi Nanopit,

First thank you for posting the error and the scenario in which it occurred. I have documented your case, in our Corrective Action Request (CAR) system under the reference number #42408. This number will allow you to track the progress of the issue as our R&D department investigates it. I noticed in the error statement that it refers to "LTR_main.vi" inside one of your LLBs, is this where you call your serial then Agilent VIs? Could you reply to this post and attach that VI?

Thanks.

Cheers,

Jonah

Applications Engineer

National Instruments

[Nanopit]

The LTR_main.vi file has many sub-VIs which it will try to load on start-up and are contained in the .llb. Should I just send you the .llb file?  Also, is there a way I can send this to the R&D department without posting it in the public forums?  This program belongs to the company I work for and has a good deal of time/money invested in it's design.

Thank you for your response and time invested. 

[jpaul]

Nanopit,

I would recommend creating a service request by going to ni.com/ask, where you will be able to directly communicate with an Applications Engineer here at NI, including sending files etc. Thanks!

Cheers,

Jonah

Applications Engineer

National Instruments