cRIO Unbound Computing Node and Deployment Challenges

[JohnG3k]

Hi All,

The immediate question I have is... what does it mean when you get a pop when deploying to cRIO that says 'Unbound computing node'? I have got this a few times as I try to debug what appears to be a corrupt vi.

The complete problem:

I have an Actor Framework based application that I have been developing for PC execution for a few months now, and runs great on the PC. I deployed it to the cRIO last sprint and worked well there as well after a few minor adjustments. I did another two weeks of development, testing on the PC as I went, and it ran well. I tried deploying to the cRIO and now get the message ‘deployment completed with error’, similar to what is described here, half way through deployment.

That article suggests a corrupt VI, however if I had a corrupt VI, I wouldn’t think it would work on the PC either? Nevertheless, that is my only lead, so pursuing that. I have started with the code from the last successful deployment, and gradually adding back functionality.

A few times in my testing I got the ‘Unbound computing node’, and would be helpful to understand what this means, as it may point me in a different direction for resolving this. Any other suggestions for debugging this would be great.

Thanks,

John

[JohnG3k]

I have found the problem... small class with three methods. After deleting the methods, it deployed successfully. Recreated the methods from scratch and now all good!

Still curious to know what 'Unbound computing node' means if anyone knows? (Google search turned up nothing).

[Jonzarwal]

Hi,

We have the same kind of problems here when tring to run a AF based app on cRIOs. It's always the same pain to run the code in LabVIEW or build and run an rtexe.

Sometimes AF gets broken while deploying (with no errors in the error window...), sometimes an error says that it can't deploy a polymorphic VI in RT, pointing to a VI where there is no polymorphic vi (but maybe refering to dynamic dispatch), and sometimes we have this incredible popup "unbound computing node"...

Without changing the code, but alterning mass compiles, clear compile cache, disconnect/restart crio, restart Labview, change dev machine, we finish with a successful launch, but without understanding what's really happeing and why sometimes it's ok et sometimes not.

Note that the same code has abosutly no problems to run on a PC.

We tried to use compiled lib (AF.lvlibp) instead of the vilib librarie, but the problem remains.

I don't have any solution, just want to say you're not alone!
And i'd be glad to have any suggestions on how analyse this. We are more and more reluctant to use NI Linux RT systems for this reason.