Log in

View Full Version : Hundreds of more watchdog errors even wih the new update


Acshi
16-02-2010, 21:49
I would have posted this onto the thread of someone who had similar problems, but found that my situation was different enough.

I have just finished installing the new Labview update, cRio image, and driver station updates.

dsVersion 10.02.08.00
cRIO Image LabView - 20010_v20

I am not sure how much is related, but am also getting many errors along the lines of "
ERROR <Code> 1026 occurred at Property Node in
NI_FPGA_Interfaace.lvlib:NIFPGA System.lvlib:Async VI Agent.vi
LabVIEW: VI Reference is invalid
"

and although probably less related, I am not able to get a battery voltage reading in the ds either.

I have not made any changes to the vision system and can get video easily, though disabling it doesn't help either.
I have not made any changes in the watchdog code for teleop either.

I get the same symptoms in both teleop and autonomous modes, and the ds indicator switches between "... Enabled," "Watchdog Not Fed," and "No Robot Code."

Any help would be appreciated.

Thanks,

jhersh
17-02-2010, 03:03
Did you rename your top level VI?

Did you run Start Communication in more than one place?

What exactly did you changes that a caused you to start getting the errors about the invalid VI reference?

Robototes2412
17-02-2010, 13:15
We have the same probelm, even though we are using java

Tom Line
17-02-2010, 13:23
I would suggest to both teams that you try loading the default framework (only changing the motor references to be appropriate). I haven't seen those errors before, so I suspect it's something you've done to your code.

Robototes2412
17-02-2010, 13:26
I've even tried running "virgin" code and get this error, its prolly a bug in the cRio firmware

jhersh
17-02-2010, 13:53
If you reimage the cRIO and run the unmodified new project you still get these errors?

Robototes2412
17-02-2010, 14:01
ya, we tried that, itll be bad if it isnt fixed before competition #ni

jhersh
17-02-2010, 14:06
We haven't been able to reproduce it... so we need you to provide more details as to what could be causing you to have this problem but not us or anyone else.

Thanks,
-Joe

Radical Pi
17-02-2010, 15:45
What does your hardware setup look like right now? Are the cRIO modules in their correct positions?

Doug Norman
17-02-2010, 16:06
Hundreds of more watchdog errors

Are those System watchdogs or User watchdogs? Can you post a copy of what gets printed to the Driver Station?

Robototes2412
17-02-2010, 22:54
it happens randomly, also being unpredictable.

Acshi
18-02-2010, 01:24
They were user watchdog errors I was getting, I didn't rename any of the default vi's or call Start Communication myself, I re-imaged my cRIO and the virgin code worked fine, so I am slowly reincorporating my code into that, and hopefully will find the breaking point soon.

Greg McKaskle
18-02-2010, 05:59
For user watchdogs, that is the right approach. As you add in the code, look for things that will take significant time in the code you are adding. Obvious things that will cause this issue are Waits, calls to vision, loops that run for many iterations, etc.

Greg McKaskle

Greg McKaskle
18-02-2010, 06:02
For the Java user, if we are done #ing NI, can you better describe your symptoms, or have you fixed the problem?

Greg McKaskle