cRIO deployment fails

[sakalaukus]

I have a cRIO-9074 and I just upgraded to LabView 2009.  

I am trying to run the latest version of my code, and the deployment fails.  The FPGA vi compiles, and I can run it directly on the cRIO, but when I try to run the RT vi, I get the following at the end of the Deployment Progress Dialog box:

Deploying jagbot_RT_v3.vi
Host LabVIEW failed to download jagbot_RT_v3.vi
Deployment completed with errors

It never actually tells me what the errors are, just that it fails to download my code.  The Deployment Progress box says there is 68% free memory left on the cRIO when it chokes.  Is there a way to determine why the download failed?

Thank you,

James

[Hueter]

What version of NI RIO do you have installed? You need to make sure that you have NI RIO 3.2.1. installed on both the development system and the cRIO itself for LabVIEW 2009

[sakalaukus]

Thanks for the reply, but I have downloaded NI-Rio 3.2.1 and made sure the software running on my cRIO is said version. 

I have tried to deploy a small example project with success.  I am also able to run my FPGA code, so I'm thinking the problem is with my RT VI.  The VI worked before I upgraded to 2009.  Its hard to fix the problem when you can't find it though.  Any thoughts?

[Ben_S]

sakalaukus,

Hello! Can you post a screenshot of the software installed on your cRIO? It may be that your missing a component. Thanks!

[Ben_S]

Are you using Scan Mode?

Also, do you see these things installed?

• NI-RIO 3.2.1
• NI-RIO IO Scan 1.2.1
• NI Scan Engine 1.1.0

[pneumann]

We have the EXACT same problem. We upgraded to LabView the other week. Our FPGA VI won't compile but after going through the NI 2009 FPGA web pages we fixed a number of issues  and we got it to compile. After testing the FPGA vi, we were ready to run our entire system. We deployed our main RT vi (called Mannequin RT 2.vi) which calls all of our sub-vis and FPGA vi. We get the same error message at the end of our deployment log:

Host LabVIEW failed to download Mannequin RT 2.vi

We have 48% memory still available so that's not a problem. Our entire deployment log is attached below. Our real-time hardware is a sbRIO type 9642. We had previously re-formated and installed NI-RIO 3.2.1 on it. We are not using scan mode. Smaller real-time VIs do deploy and run normally but not our large main vi (Mannequin RT 2.vi) which controls our entire project.

Here's what is installed on our sbRIO. Any ideas what is going on? We have a deadline at the end of the month for our project and right now we can't even run it.

-Paul

[Hueter]

Here are a few things that you can try to narrow down the source of the problem

1. Can you compile and successfully deeply a shipping example. 

               This will tell you if there is a problem with the software settings, application or network communication

2. Can you FTP the large VI on the the controller after it is compiled and run your application? 

               This will tell you that the code fits and that it compiled correctly.

3.You can try to recreate your LabVIEW project to make sure you didn't do anything odd in the settings.

               This will tell you if you set up your project in an odd way.

Let me know how this goes.

[Jennifer_R]

Hi James and Paul,

I saw a similar issue recently.  In that case, it seems part of the main RT VI may have been corrupted during the upgrade process.  (When I mass compiled the directory containing the project code, the main VI was listed as a "Bad VI.") 

I was able to determine which portion of the code caused the problem by removing various parts one-by-one and trying to run the VI.  Hopefully you can then recreate the problematic section and run without any errors.  If you isolate the cause and still can't run the VI after deleting and recreating the code, try placing that section in a subVI. 

Hope that helps!

[Jennifer_R]

For anyone who is following this thread, this was reported to R&D (# 186326) for further investigation.

[pneumann]

Just to give everyone an update, I tried a number of suggestions to get our main RT VI to deploy without any luck under LabView 2009. Again our main RT VI would error out when we deployed our application to our sbRIO with a  "Host LabVIEW failed to download Mannequin RT 2.vi" deployment error message.

By chance after running our main RT VI through VI Analyzer Tool, it mentioned the debugging property was turned on under the VI properties. Turning the debugging property off somehow corrects the situation - the VI now deploys and our entire application runs.

I have no idea why this works. To be honest,  I still need the debugging option, but right now turning it off is our only option to get our application to deploy.

Looking at Total Memory Usage (from the same VI Properties):

With      Debugging ~5551K

Without Debugging ~5498.6K

It doesn't seem to be a significant difference - (unless there's some sort of 5500K VI size limit?) Again we have still 47% of our sbRIO memory free so we're definitely not running out of memory on the sbRIO - in fact we have plenty.

Again, I don't why turning debugging off would make our main RT VI deploy but it does.

-Paul