cRIO crashes when loading code

[Joe Ross]

We're having a problem where 75% of the time that we try to load a program, it is unsuccessful because communication is reportedly lost. This is independent of programming computer, cRIO, loading method, and connection method (wireless/wired).

Initially I thought it was connection problems, however, after looking at netconsole, it looks like something on the cRIO is crashing. See the attached netconsole log.

I used the system manager when my program is running and it shows CPU load at approximately 95%. However, if I remember correctly, system manager has a decent amount of overhead, so the usage might be ok.
blem
I tried a few times with the no app switch set, and loading the software was always successful, however I haven't run it enough to be convinced that solves the problem.

[Greg McKaskle]

The cash log doesn't have as much information, but looks quite similar to the stack overflow issue that netconsole had with v27, thus the release of v28.

If may be useful to disable netconsole and see if it goes away. Also, what cRIO version?

Greg McKaskle

[Chris Hibner]

We have the same issue as Joe, with netconsole turned off.

Our CPU load should be pretty low, we disable 95% of our PeriodicTask code while in disabled mode (we did that specifically to try and avoid this problem). We still have issues. I've been meaning to look into it further, but we have more pressing things to work on with ship day coming.

[Joe Ross]

This is with v28, and occurs with netconsole off also.

[Greg McKaskle]

Would it be possible to get a project zip to Doug or myself?

Greg McKaskle

[Joe Ross]

Email sent to Doug and Greg

[Joe Ross]

Doug was not able to reproduce the problem with my project.

We did look at the processor utilization. When I reduced it to 60%, the problem still happened. When I reduced it to 40%, the problem did not happen.

At this point, I'm using the noapp switch to do all of my code loading.