Code Not Deploying

[cdizzle]

Hi Everybody,
So my team has been experiencing some odd problems lately.
We are trying to deploy our code to the cRIO from labview, so we attempted selecting "build all" and then selecting "deploy all," but what happens is that it says that the code is deployed successfully, but on the driver station, the robot code light stays red.

Any ideas on what could be wrong?

-Team #2367

[Mark McLeod]

Use "Build", then "Run as startup"
from the Build specifications -> FRC Robot Boot-up Deployment


"Deploy" only copies the code to the cRIO, but doesn't tell the cRIO to run it.

[Cecil]

Instead of "Deploy all", you need to use "Run as startup". Keep using the "build all", but to get it to run permanently, you have to use the run as startup command.

[cdizzle]

Ok, I'll try that, thanks for your help!

[cdizzle]

Originally Posted by Mark McLeod

"Deploy" only copies the code to the cRIO, but doesn't tell the cRIO to run it.

Wait, then why would you want to only copy the code and not run it?

[Mark McLeod]

The CRIO is a commercial product.
Look on your own computer at the multitude of programs that have been installed (deployed) to your disk drive, but don't run until you ask for them.

In the commercial world the same holds true for the cRIO. Some applications will put several programs onto the cRIO to be used only when they are asked for.

[Greg McKaskle]

I know, I know. That menu has a lot of lower-level choices that really shouldn't be at the same level as the Run as Startup. I'm trying once again to get it changed, but the RT team always seems to have bigger fish to fry. Actually, comments like yours help.

Greg McKaskle

[cdizzle]

Hi guys,
This is a little embarrassing, but I'm not able to find run as startup when I right click Build Specifications

UPDATE: Never mind, found it. Thanks for all your help guys, sorry for the inconvenience!

[cdizzle]

Hi guys, I think we have a new problem, the labview software cannot find the cRIO when deploying. The IP address is correct, and the driver station and the imaging tool can find the cRIO, but when I try to load the code, it keeps on "waiting for the cRIO to respond." It also sometimes says that it "lost connection to the cRIO"

[Greg McKaskle]

There was a bug in the RT runtime that kept LV from exiting out of deployed applications in order to start another. The LV 5.1 update works around this. The workaround until you install the update is to exit the dashboard, then deploy code, then launch the dashboard.

Greg McKaskle

[cdizzle]

Originally Posted by Greg McKaskle

There was a bug in the RT runtime that kept LV from exiting out of deployed applications in order to start another. The LV 5.1 update works around this. The workaround until you install the update is to exit the dashboard, then deploy code, then launch the dashboard.

Greg McKaskle

We tried exiting the dashboard and then running the code, but the cRIO still isn't responding.

[Greg McKaskle]

Still isn't responding means it is waiting for ...

Also, the RT bug I described would peg the CPU when it was waiting for the deployed code to exit.

The bug would occur when any SmartDashboard client was connected, I mention the dashboard because that is typically the only one teams run, but if you are running Smartdashboard, Table Viewer, RoboRealm, or other Smartdashboard clients exit those too. And better yet, install the LV5.1 update and the bug will go away.

Greg McKaskle

[cdizzle]

It worked after I re-imaged the cRIO, if that helps in any way.

[Greg McKaskle]

If imaging the cRIO involves formatting it, then the deployed app will be deleted. So it is the other, much longer, workaround.

Greg McKaskle

[cdizzle]

Ok, I'll try updating the LabView code, thanks for your help!