Cannot deploy to Roborio

[dmay31]

Our team is having problems deploying code via LabVIEW over USB. We can ping the rio with the IP address and the .local address with USB connected. We can also access the web view page at the 172.22.11.2 address across USB. Trying to deploy always results in a timeout. Any help would be greatly appreciated!

[dmay31]

More details. All this was done with the laptop connected to the Roborio via USB.

1) We reset the Roborio to factory defaults

2) Updated the Firmware with the FRC Tools 4.x

3) Used a non-school laptop that has the firewall disabled

4) Was able to deploy our code from LabView the 1st time

5) Five additional attempts to deploy failed

6) Rebooted the Laptop

7) Deploys still fail

😎 Check we could ping the Roborio with the IP address and the .local name

9) Checked we could SSH into the Roborio and it worked

10) Tried to deploy again and it failed.

I know our LabView code has some console prints in it that we were using to test the color sensor. Is it possible the prints are interfering with the deployment? Can we disable the startup program in the roborio that we have running to see if it is preventing the deployment?

[oscarfonseca]

When you say that the deployments results in a timeout, what do you see in the deployment window? Something among the lines of "waiting for the target"?

If the deployed application is consuming all the CPU (for example, with loops that have no timing bounds) then it is likely the issue is in the realm of what you're exploring. You can disable the startup application using the imaging tool: Edit Startup Settings > Disable RT Startup App. You can also accomplish this through NI MAX.

Cheers,

[dmay31]

Thanks Oscar!

However...

1. We swapped the roborio and now we can deploy.

2. Now the robot code will flip on and off for a couple of seconds before we can enable.

3. Is there any other helpful information/logs that we could get off the roborio or the computer and if so how can we get it?

[oscarfonseca]

Thanks for providing the additional details.

1. We swapped the roborio and now we can deploy.

• Interesting. I find unlikely that the roboRIO itself would have an issue or behave different to another one.
• That being said, you always have the option to set the "faulty" one to safe mode by pressing the Reset button for ~5 seconds and making the STATUS LED to be blinking 3 times. Then use the imaging tool to reimage it (it will erase any startup application and configurations loaded on it).

2. Now the robot code will flip on and off for a couple of seconds before we can enable.

• It may be an issue with the code's initialization, then. I recommend taking a look at how you are initializing the different resources (motors, sensors, etc) in the code.
• If you are using CTRE devices (Falcon, Talon, CANivore), please look at the published errata and update their software.

3. Is there any other helpful information/logs that we could get off the roborio or the computer and if so how can we get it?

• I would start with the logs that are in the driver station. Access it through the small gear on top of the text outputs on the right of the driver station. You can learn more here: Driver Station Log File Viewer 
• I would look for the CPU load first, and then the events tab to find any errors.
• There are other logs that get stored in the target (the roboRIO) but I don't think those would be as useful in this situation.

Cheers,

[dmay31]

Oscar,

Thank you again for your help. We had a team meeting tonight, but I didn't see your response in time to give them a try. I'll give them a go at your next meeting.

Unfortunately, we again started have problems deploying code. This time I installed Wireshark on the computer and sniffed the traffic on the interface to the Roborio.  In traffic I can see what appears to be the phoenix turning getting JSON data back on port 1250. When we try and deploy code I can see the computer trying to open a TCP connection on port 2343, but the Roborio resets the connection. I've attached the log if that helps.

Thanks

Dale

[dmay31]

Thank you! Deployment is working consistently again.