No Robot Code EVER -- Help Please

[cjlane1138]

Hello everyone,
So, my team was programming our robot. One minute everything was working fine, and the next, I deployed, and we started getting this huge "No Robot Code" problem.
We were actually able to drive the robot at one point. Then, it stopped working all of the sudden.
Here is what we did in order:

Build->Deploy->Reboot->No Robot Code
Reboot->No Robot Code
Reboot of Classmate->No Robot Code
ReImage of CRIO->Rebuild of Code->Deploy->Reboot->No Robot Code
Build->Deploy->Reboot->No Robot Code
Build->Deploy->Reboot->No Robot Code
ReImage of CRIO->Rebuild of Code->Deploy->Reboot->No Robot Code

This entire time, we were connected to the robot through the wireless bridge. We were getting comms every time, but No Robot Code.

Is there anyone that has had the same problem and fixed it? Or is there anyone that has suggestions on how to fix it possibly?

If there are any more details you need, just reply with what you want to know.

Thank you,
Eagle Engineering 1138

[ianonavy]

Check to make sure that your cRIO modules are in the right slots and that the cables connecting them to your digital sidecar have a proper connection. If you want to be really ardent, you can try checking continuity between each pin using an ohmmeter. You might want to ensure that all of your wires are connected properly and that your sidecar is receiving adequate power. Our team had a similar problem and replacing the sidecar ended up getting the Robot Code light to finally turn green.

It's more than likely a wiring issue than a programming issue from experience.

[cjlane1138]

Originally Posted by ianonavy

Check to make sure that your cRIO modules are in the right slots and that the cables connecting them to your digital sidecar have a proper connection. If you want to be really ardent, you can try checking continuity between each pin using an ohmmeter. You might want to ensure that all of your wires are connected properly and that your sidecar is receiving adequate power. Our team had a similar problem and replacing the sidecar ended up getting the Robot Code light to finally turn green.

It's more than likely a wiring issue than a programming issue from experience.

Thank you. Although, we are not really sure whether it is a wiring issue. We heatshrinked, wrapped, and labeled everything. Everything is neat. But in the morning I will have our electronics lead check it out! Will let you know.

Thank you!

[Mark McLeod]

If you are using C++ or Java, then fire up Netconsole to see why the code is crashing.

If you are using LabVIEW then you should be doing:
Build
Run as startup

[cjlane1138]

Originally Posted by Mark McLeod

Build
Run as startup

We have done this, many times

[Mark McLeod]

Try running in debug mode to see if something happens to the code as you watch it execute.

[Mark McLeod]

Originally Posted by Mark McLeod

If you are using C++ or Java, then fire up Netconsole to see why the code is crashing.

What did it show?

Originally Posted by cjlane1138

Sorry everyone, I forgot to tell you guys that we are coding in C++, not LabView.

Post in the C++ sub-forum next time and you'll get more help.

[Greg McKaskle]

My guess is that the code is crashing very early in its execution. The Robot Code LED is based on getting data from the user framework code via the FRC_Communications.out file.

As others have said, you'll get more help from the C++ forums, but I'd try to use the debugging tools, even just placing print statements every five or ten lines to see how far you are getting will help. STOBbing is another technique many developers find helpful. It stands fore Sync To Old Builds. You use source code control to return to previous builds and determine if the problem existed there. Then you diff the changes you've made in a few at a time to identify the culprit. Personally I find it quicker and more exciting to debug, but both are ways to make progress.

Greg McKaskle

[lscime]

Make sure all of the DIP switches on the front of cRIO are in the correct orientation (one of them may have accidentally changed since the time when you last saw your robot working). In particular make sure that the "No App" switch is in the "off" position.

[androb4]

What you're supposed to do is:
1) Go to the project explorer in LabView
2) Expand the "Robot Specifications"(I believe) tab at the bottom of the list.
3) Right click one of the items on the list that'll show. Build, and Set/Run As Startup
4) First Build it.
5) Then Set is a start up.
Done Deploying!!

[androb4]

Did you make sure that you have the most recent updates. And that you imaged the cRio correctly??

[cjlane1138]

Sorry everyone, I forgot to tell you guys that we are coding in C++, not LabView. So we are using WindRiver. And yes, we have the most recent updates and we reimaged the CRIO 2x correctly.

[cjlane1138]

Also, I forgot to say that the robot was working before it started giving us "No Robot Code"

[saphiren]

You're not the only team to have this problem, check this thread to see how 3841 got their code to work (the most helpful stuff is close to the bottom).

A few things to check: is your DB37 ribbon cable one of the ones that shipped with the backwards connector, and did your team fix it exactly as specified in FIRST's guide? Have you gotten the code to run before, or is this the first time you've tried deploying it? Have you been deploying the correct .out file to the robot?

Hope this helps