9004 startup problem

[Olaf_Stetzer]

Hi all, I have a cRIO 9004 system with various I/O modules attached. I have a RT startup app that used to run properly but after some changes and re-programm of some components the program does not boot anymore. The strange thing: It runs perfectly when I "deploy" it from my project (which means I start it with the UI enabled so that I can debug the program while it runs). However, when I build it and set it to "run as startup" and reboot, nothing happens. I checked all possible "easy" mistakes like dip switches and config file settings, IP settings, all seems find. I have LED blinking in a loop which dies not blink. There are only few things happening before the loop is entered: Read a config file /using open g libraries) and configure RT FIFO's and queues. Is ther something that could go wrong here (e.g. file path settings) even when the program does start "remotely"? Thanks!

[Himanshu_Goyal]

Hello Olaf,

First of all I just want to say that OpenG libraries are not fully recommended by National Instruments for Real time System. I think problem is related with your building EXE and set EXE as startup. Try building the EXE again and keeping your OPEN G library folder in Source file always include section.

As you mention that its working fine from your project but not working in RT startup. As you are reading the file in the RT make sure file is available in RT on same path or give constant path in RT. May be you are using the building the path function and in RT EXE its different. So for checking gives constant path.

[Olaf_Stetzer]

I doubt that the open g functions are the problem. They used to work in the past years in the same way I am using them now. However, I do remember that I had trobles recently to set the path to the config file in the RT app correctly. I have a control on the RT app to set the path, and it was difficult to figure out what path I need to use, do I need a C:... path or something else? The Path control however did not accept the path without drive letter although I think this is what I thought would be correct. Maybe I am mixing things up now, I write this from my memory. But is it possible that this is the problem?

[Olaf_Stetzer]

I checked now again: The path to the config file is set to: c:\config.ini If I click on it I get the file folders on the laptop where do the programming, not the RT target. I do have other paths on the rt app which are also soet to c:\.. something for data logging and there the new log files are created on the RT target in a folder called C which puzzled me a bit. For the record: I use LV 2011 (service pack 1) and RT 4.1 so pretty recent.

[Himanshu_Goyal]

Hello Olaf,

You have to check the file folders in the RT, FTP the RT target and check the file are available on same path or not....

[christian_w · Accepted solution]

So after the reprogramming it might be that you forgot to refresh the build specification and one of the new VI's is missing. Try to check or setup a new build specifcation.

Hope it helps

Christian

[Olaf_Stetzer]

The config file is in the root folder on the RT target. I now changed the path control to a string control and use the string to path function. The path I use is "\config.ini" and it still works in "remote mode", so with UI on the laptop. The config file is read (checked with probes) and everything works fine. If I build and donwload the vi as startup app and reboot, nothing happens. Could of course also be something else and unrelated the the config file/path....

[christian_w]

If you build the RT executable with debug information you will be able to connect and debug the running application. Try that next...

[Olaf_Stetzer]

---

@christian_w wrote:

So after the reprogramming it might be that you forgot to refresh the build specification and one of the new VI's is missing. Try to check or setup a new build specifcation.

Hope it helps

Christian

---

Thanks Christian,

that was actually the problem. I deleted the old build specifications and generated a new one and now it works.