LabVIEW 2015 SP1, cRIO-9075, Restart command doesn't restart, hard reset required.

[Drewski]

Hey All,

Early on in discovery here. I compile the RT, and then deploy the compiled startup app to the cRIO successfully. When I do a restart from the utility menu of the project, the cRIO goes away and doesn't come back until a hard reset.

Same behavior from within the RT code. The RT code calls the restart vi and the cRIO goes away and doesn't come back until a hard reset.

Anyone else have experience with this issue? What is the solution to be able to restart through the project or the RT code.

Thanks,

-Drewski

[elcalverado]

Drewski,

Early on in discovery here. I compile the RT, and then deploy the compiled startup app to the cRIO successfully. When I do a restart from the utility menu of the project, the cRIO goes away and doesn't come back until a hard reset.

What do you mean by "goes away and doesn't come back"? Do you mean the cRIO disconnects from the LabVIEW project? Or are you saying that the cRIO doesn't boot successfully until you hard restart it?

-Jordan

[Drewski]

CRIO is disconnected and doesn't come back until a hard restart.

[elcalverado]

It's normal for the cRIO to disconnect from the LabVIEW project when you restart it. The restart shuts down the network communication on the cRIO, so whatever it is talking to is going to see it stop communicating. After the restart, you would want to give the cRIO time to boot up completely before trying to reconnect to it. 

-Jordan

[Drewski]

The Real-Time code sets up the watch dog for 5 seconds and then in a loop that runs every 100 ms, checks a Network Shared Variable (string) for matches to a case statement and whacks the watchdog. If there is a match, the RT writes to the NSV. The lights on the cRIO, solid green, flashing Ethernet lights, duration of 5 minutes make me think the cRIO has booted up completely, yet doesn't update the NSV in response to the host.

Now when I press the reset button, the same sequence occurs, solid green light on cRIO, Ethernet lights, duration of 5 minutes, and the NSV updates in response to the host as expected.

So why is there a different behavior between hard reset and soft boot? How can I get the hard reset behavior from within the RT code?

[elcalverado]

Generally, I wouldn't expect there to be any difference between a hard and soft restart. You rt executable is set to run at start, right?

Do you have any error or status logging so that you can ensure that the cRIO is booting into the rtexe successfully after restarting? You could write a timestamp to a log file as soon as the rtexe starts to check this. I would also recommend implementing some kind of error handler/logger so that you can go back and see if any errors are occurring that are preventing the watchdog variable from updating.

Another simple method of checking that the RT code is running is to blink the user LED on the cRIO.

-Jordan