What is the root cause of "...loaded with errors on the target and was closed" error when deploying VIs to cRIO

[Petru_Tarabuta]

In reply to Quiztus: Thanks! "now it is NI's turn to hire a second developer for LabVIEW or let the one stop debugging BD zoom" - Haha.

In reply to Steffen: Sorry, I don't fully understand the suggestion. The issue above occurs when deploying to the cRIO's volatile memory. In other words, it occurs when simply running (Ctrl + R) the top-level VI on the cRIO. I have not yet tested whether the same behaviour occurs when building an rtexe (i.e. deploying to non-volatile memory).

[Intaris]

---

@Quiztus2 wrote:

I think this non-deterministic behaviour can only be corrected by NI. You did a perfect job by documenting this, now it is NI's turn to hire a second developer for LabVIEW or let the one stop debugging BD zoom.

---

Things I've learned are connected to this problem occurring:

Separate compile code

Inlined VIs

Classes

Disable structures

Conditional Disable structures

Typedef changes

Phase of the moon

OK, the last one is not serious. But it seems that the sequence of actually loading files on different targets tends to skip important steps or mixes up some important infos (Like symbols set tor conditional disable structures). All of the other things seem to lead to some kind of invalid loading from cache. The reason why I say it because clearing the compile cache nearly always solves my issues. It's not a nice solution, but it's at least something to help get me working again.

[Petru_Tarabuta]

One more data point: My colleague and I have successfully deployed and ran two cRIO Main VIs today. We deployed and ran the VIs a total of 5 to 7 times. The deployment never failed.

We are now using a single cRIO target per project, and no files under the My Computer target. We are also using the "useCacheForDeployment=False" key in the LabVIEW.ini file. I can't be certain that one or both of these changes fixed the issue, but the deployment has been completely reliable in the admittedly short time (a few days) since we applied these changes.

We are continuing to use "Separate compiled code from source file" in all our files - we did not change this. We have not done any Mass Compile or Force Recompile in the last few days.

[Quiztus2]

I noticed the release of compactRIO 24 Q4. But there are no release notes.

NI CompactRIO Release Notes - NI

[Petru_Tarabuta]

Another data point: My colleague and I have successfully deployed and ran a few cRIO Main VIs today. We deployed and ran the VIs a total of around eight times. The deployment never failed. We are using the exact same configuration as yesterday. In short, the deployment still works reliably, thankfully.

[Petru_Tarabuta · Accepted solution]

My colleague and I have continued being able to successfully deploy and run cRIO VIs on multiple occasions in the last few days. The deployment has become completely reliable since we started using a single cRIO target per project (no code under My Computer) and using the useCacheForDeployment=False token in the LabVIEW.ini file. Therefore, I am marking this as the solution. This is a workaround, as the problem should not be happening in the first place.

We have not unticked the "Separate compiled code from source file" setting, and we have not investigated the cRIO Error Logs in any depth.

Many thanks for all the help!

[Intaris]

I don't want to burst your bubble, but we've observed the same thing. We change some random thing and the deploys start working again.

This remains the case for days, weeks or even months and then it just stops working again. And of course the steps we took LAST TIME to fix it don't work this time. This is why I put "the Phase of the moon" in my original response. It was meant as a joke, but the reason why I found it funny is because it's something which has actually occurred to me in the past.

I'm interested in whether the problem rears its head again in a month or two.

[Petru_Tarabuta]

Thanks. I accept that it could start happen again. I will keep an eye on it and update this thread if/when it happens again (hopefully never!).

[Quiztus2]

I don't use the .ini entry anymore, since I switched to 24Q3.
So far I had no deployment issues, that couldn't be solved by clearing compile cache with this.

fingers crossed