Hello Kurt and Matt,
After a legnthy process, here is what I found-
Prior to installing the GPIB-USB-HS drivers, I had installed some Hewlett-Packard/Agilent IVI/VISA Drivers for another instrument. I found that I had to uninstall these drivers to get rid of the code 39 errors. Apparently the Agilent drivers (of the same name "visa23.dll") will check to see if they are still OK. when they detect they have been overloaded, they rewrite themselves from cache I think. After I completely unistalled the Agilent VISA software, I was able to avoid the code 39 error.
I have pasted in a partial transcript of the solution thread with hopes that helps or at least points you in the right direction(starts at bottom but solution is last couple of e-mails):
Ben,
BTW, I had one other machine that had that HP package on it and that
would give me some problems too on installation of one of my apps that
made use of the VISA packages. When I would start up the app, it would
tell me that it couldn't find visa32.dll but of course it was right
there in the system32 directory. Maybe since the machine I was
developing on had the problem with the conflicting software and I would
build the app on that machine that it would pull the wrong DLLs and put
them in the setup and install package. Alternatively, it could be like
you said too, that after installation, the Agilent package would
overwrite the NI DLL. I guess I can't have the two packages on the same
machine at once. Also, as a help for future debugging, when I was
removing the Agilent packages from the other machine I saw the name of
the one remaining one that I had to remove before the NI-PAL would
start. It was called the "IVI Shared Components" in the windows removal
list. If you get another code 39 problem tell them to look for that one
too. The Agilent 34970A install puts at least 3 different items in the
uninstall list and that's one of them.
Best regards,
Andy
-----Original Message-----
Note: Your reference number is included in the subject field of this
message. It is very important not to remove or modify this reference
number, or your message may be returned to you.
Andy,
You are welcome for the support. I offered it to you becasue I had seen
the effort that had gone into the GPIB error 39 in the past without a
solution. In the future, Agilent IVI software will be something we look
for with installation issues involving these devices. Thanks for
providing the detailed information about your Agilent software. We use
these kinds of notes from past service requests very often.
Regards,
Benjamin
Applications Engineer
National Instruments
http://www.ni.com/support
------------------------------------------------------------------------
------
Thank you for your unsolicited support. I hope this knowledge can help
to solve at least the one other individual that had the problem. And
maybe more if the right questions get asked. To review the history, I
was using an rs-232 connection to an Agilent 34970A and had loaded their
software (agt34970A_XL.exe) to capture some temperature and voltage
data. That had the IviSharedComponents1.3.0.msi installer as part of it
apparently. It installs as several parts that show up in different
places on the windows uninstall list, so when I uninstalled it,
apparently I missed that piece. I notice that there are still dll pieces
in the system32 directory though. Do you think they may conflict? One
that I notice is "visa32.Agilent Technologies.dll". When I examine the
file properties, it's "product version" is 14.2.8931.1 but the "file
version" is Version:78.2.8931.1. The "internal name" and "original file
name" is visa32.dll. If anybody uses those properties to identify the
file, I'm sure that would screw things up. The other piece that's
obvious is "visaConfMgr.dll" (visa conflict manager) from the IVI
foundation but that name is less conspicuous. I'm not sure what product
installed that one.
On the persistence (stubbornness), Yes I admit to it. I've been working
on computers since the 70's- the first I built was a intel 8008-1 at a
screaming 800 kilohertz. They are all the same - no matter how smart you
allocate the memory or try to hide the software innards or how many
cpu's and instruction que's you have. It just gets more messy, faster.
Regards,
Andy
-----Original Message-----
Note: Your reference number is included in the subject field of this
message. It is very important not to remove or modify this reference
number, or your message may be returned to you.
Andy,
That is awesome. I would agree with you that the problem was the
Agilent IVI drivers. They use common components with some of our
software (GPIB,
VISA) and could easily cause a conflict. In the past, we have seen
problems with software called Agilent I/O Libraries v14.1. That
software runs a watchdog service that constantly monitors some dll's.
If we install our driver and it overwrites one of the common components,
their service immediately replaces it with theirs. They might be doing
something similar with IVI now. I have to say I'm a bit surprised that
you figured this out
- it's not often that I see this level of effort on the part of
customers.
It's great that the GPIB-USB-HS now works in your computer.
If all NI processes are no able to start correctly, there is no need to
perform the steps I suggested. Those steps were only to help start the
NI-PAL Service Manager (nipalsm) process. Thank you for continued and
concentrated effort on this issue.
Best regards,
Benjamin
Applications Engineer
National Instruments
http://www.ni.com/support
------------------------------------------------------------------------
------
Ben- Good news! After clearing the machine with msiblast then installing
my downloaded copy of ni488.2 v2.7 as I indicated that I would do in my
previous e-mail, I am now able to insert the GPIB-USB-HS, the driver
loads and MAX communicates with a device I have connected to the
interface using VISA. Eureka! And I didn't have to wipe my disk!!! Yea!
Ok-I can still try your suggestion and I will but now I've got work to
do. Although I'm not sure that I can tell if it will do anything. My
guess is that the IVI library was the culprit at this point. What say
you?
Thank you for your support,
Andy
-----Original Message-----
Note: Your reference number is included in the subject field of this
message. It is very important not to remove or modify this reference
number, or your message may be returned to you.
Andy,
I have an idea that might help in allow nipalsm to run. By the way,
nipalsm is the NI-PAL Service Manager and it is required as it helps
load other programs.
1. Close all NI programs such as MAX, LabVIEW etc.
2. Browse to C:\Documents and Settings\All Users\Application
Data\National Instruments\MAX\Data (you will need to show hidden files
and folders to see the Application Data folder) and clear all the
contents of this directory.
Sometimes, you will not be able to clear all the contents of this
directory. In this case, open Task Manager and end all NI related
process.
Then, try clearing the contents of Data directory.
Tip: All NI related processes in Task Manager start with nixxxx, as
shown in the image below.
3. Restart your computer.
After you do this, see if you can get the NI-488.2 Configuration Manager
to load from services.msc.
Regards,
Benjamin
Applications Engineer
National Instruments
Message Edited by Support on 10-20-2009 05:15 PM
