Issue deploying robot code to cRIO

[Rusttree]

We're having a strange problem with deploying code to the cRIO. We're a 2nd year team and we did not have this same issue last year.

When the cRIO and the development computer are freshly rebooted, everything works fine. We can build the code and deploy it with no problem. The driver's station connects and we can drive the robot around. If we, say, make a change to the code, rebuild, and redeploy, the deployment process hangs at the "Waiting for real-time target to respond" stage. It's very repeatable.

Here's the sequence that causes it to hang:
1. Turn on the driver's station laptop and open Dashboard.
2. Reboot cRIO and development computer.
3. Right-click on Robot Main.vi in the project explorer window and select "Run". Code deploys with no problems.
4. Reboot cRio when prompted.
5. Wait for Dashboard to connect and press the Enable button. Verify the robot running the newly deployed code.
6. Press Disable on the Dashboard.
7. Redeploy by right-clicking on Robot Main.vi and selecting "Run" again.
8. Hang forever on "Waiting for real-time target to respond".

Here's a work-around we found that seems to implicate the Dashboard. After steps 1-6 above, we reboot the cRIO and immediately close the Dashboard program. After the cRIO boots, we can then deploy the code without a problem. We can then reopen Dashboard and everything works fine. If Dashboard is running on the driver's station laptop during the redeployment, it will hang. It seems that once the cRIO connects to the driver's station and the Enable button is pressed the first time (steps 1-5), something about Dashboard is preventing the cRIO from reconnecting and accepting a new program after that.

Our team number is 4146, so our IP addresses are:
cRIO: 10.41.46.2
Driver's station laptop: 10.41.46.9
Development computer: 10.41.46.6

Has anyone else experienced this problem? Can anyone verify that it does or does not happen to them when they follow the same steps above? Any troubleshooting ideas are greatly appreciated.

EDIT: Another point of interest. We're using our cRIO from 2012, but we successfully flashed it with this year's (2013) firmware. We also successfully flashed the DLink using the new utility.

[Greg McKaskle]

I verified that there is an issue with RT connection.

If the cRIO has a deployed app and SmardDashboard communications are under way, an attempt to connect a LV project to the cRIO will hang "Waiting for ...".

This doesn't occur if SmartDashboard isn't in in use, and doesn't occur if the Dashboard isn't open.

Back to your post, Running a VI via the right-click you are using, or opening the panel and clicking the run button and running the VI in the debugger should never prompt for a restart. A restart is needed only when you Deploy or Run As Startup.

So, I have enough information to report and debug the RT issue. You shouldn't be bothered by this provided you run in the debugger. I'd like to understand what step is causing the prompt for reboot.

Greg McKaskle

[Saberbot]

Greg,
I was working with rusttree during this troubleshooting. I don't remember there being a prompt for reboot when we ran the code directly from robot main. It only prompted when we set it as startup. However, we ran into this problem both when we did a run and a startup deployment.

Since you believe this is related to smart dashboard/netbook tables, should we try this test again, but only close the dashboard instead of closing the entire driver station? That could be an OK workaround until we have to start working on our dashboard.

[Greg McKaskle]

Yes. I believe this only happens when running a dashboard that is connected to the robot code and having a deployed app already loaded on the cRIO.

When I had no deployed app and was simply running VIs using the arrow or Run menu, I didn't see the problem.

Greg McKaskle

[Rusttree]

Thanks for taking the time to look into this, Greg. Your help is very appreciated. We'll give it a try tonight or tomorrow and see if it works for us.