Trouble with deploying program

[sparks-1784]

Hi. We've been having some difficulty in deploying programs to the cRIO. We updated cRIO with the latest image and had deployed a program but it didn't appear 'take'. Subsequent attempts to program showed a communication timeout with the cRIO.
We then attempted to re-image the cRIO and had the following error:
Unable to complete operation because of unexpected error:
Error 56 occurred at TCP Open Connection in NI_InternetTK_FTP_VIs.lvlib:FTP Open Data Connection.vi->NI_InternetTK_FTP_VIs.lvlib:FTP Data Send.vi->NI_InternetTK_FTP_VIs.lvlib:FTP [STOR].vi->ConfigUtility.lvlib:2012_FTPSubFolder.vi:1->ConfigUtility.lvlib:2012_FTPFolder.vi->ConfigUtility.lvlib:ReImageTarget.vi->ConfigUtility.lvlibialog.vi

Possible reason(s):

LabVIEW: The network operation exceeded the user-specified or system time limit.

Not sure what the issue is here. One possible problem is that the framework we used to create our program may be compromised (just a guess). Instead of re-drawing the program from scratch, there is a strong 'desire' to copy bits of last years program for expedience sake. I noticed that when this was done, the compressor commented code in the periodic tasks vi was not in the code at all (commented or otherwise - and no, our compressor is not working because the control loop stuff is not in the periodic tasks vi).

Can anyone provide some hints here? Trying to load code into the cRIO has never been more difficult than this year and I have to think that we have a basic configuration problem.

[Mark McLeod]

Are you doing "Run as startup" (NOT "Deploy") when you tried to download your code. That can prevent it from taking.

The PC you're using:

• IP address matches the cRIO (e.g., team number)
• All other network interfaces Disabled
• All firewalls off
• All anti-virius programs off
• Ethernet connected to a switch (DLink preferred), connected to the cRIO?

You can try putting the cRIO in Safe Mode and reimaging with the cRIO Imaging Tool (a special sequence will occur), then turn Safe Mode off, reset the cRIO and try imaging again. All followed by putting your code on the cRIO using "Run as startup".

[sparks-1784]

Thanks Mark.
We were selecting 'run as startup' during all the attempts to reprogram.

We put the cRIO in Safe mode and ran the re-imaging utility - there was some strange behavior in safe mode when we did this - safe mode was not indicated on the imaging utility after rebooting the cRIO. The first time we installed the 2012 image, things worked ok. Subsequent attempts have been 'painful'.

We will disable the other network interfaces (didn't do that, wireless was still running)
The ethernet was connected to the d-link (I think). I normally connect to the cRIO directly. We are using the crossover cable for reprogramming.

Turning off the firewall on the PC is a good idea.

We will go through the whole thing again today. Originally, we didn't have any trouble with it, but for some reason - recently - we started having issues.

[sparks-1784]

So - to close this issue out (for the benefit of others)...
We were able to get the cRIO to program every time without the need for periodically running the imaging tool.
The problem appeared to be that the incorrect robot framework was being used - last years framework was used which apparently causes issues with the 2012 cRIO configuration.
A 2012 framework was opened and elements that were desired for re-use were copied from the older files and pasted into the new framework. I thought this was done initially, but apparently the old framework was used instead.
I personally encourage team members to redraw designs that are being re-used in a new framework. One can gain experience using labview through the re-drawing process.

[Mark McLeod]

Thanks for the follow-up.

[Alan Anderson]

Originally Posted by sparks-1784

I personally encourage team members to redraw designs that are being re-used in a new framework. One can gain experience using labview through the re-drawing process.

It's not just a good idea -- it's the law.

Seriously, the robot rules can be construed as prohibiting the use of software that was written before the start of robot build. It's akin to reusing a fabricated mechanical assembly from a previous year's robot. Hoever, making the software available in a widely-recognized repository can turn it into the equivalent of a COTS component, so it won't be considered illegal.