viSci

For some reason I had assumed that the top level actor waited for child actors to reply to the boilerplate shutdown message. I will use your suggestion in the future, thanks!

I use subpanels frequently for UI management of instrument actors in a large SCADA application. Everything works well until shutdown of the top level actor. I am seeing some evidence that the child actors running in subpanels do not shutdown gracefully. For example in the cleanup case of my subpanel actors I have the following code which attempts to save the actors control settings to an ini file. This works fine in the IDE but fails in a built exe. As far as I can tell the vi reference is still viable at the start of the Cleanup case but the ini file does not get written to. Subpanel references are handled by the messenger framework automagically so I am wondering if there might be something going on I need to consider. Also I added the Save Default Profile message which I can call from the top level prior to shutdown and that seems to work in the exe.

Are you suggesting that the problem was non-uniform application of the pointer-sized integer CLFN parameters?

It crashes running any of the included examples. It seems to be able to open a reference to the .db but any queries crash.

I did try the latest sqlite dll from SQLite.org and also adding the AES symbol to the project but still crashes.

Thanks. I have already started porting to JDP SQLite. I should have done that a long time ago.

That is considerably more modern then my v1.3.1, I would give it a go if you could make it available...

I realize this SQLite LIbrary (by ShaunR) is pretty ancient so not surprised there maybe some dll updating required but I thought I would inquire. I have used it many times in the past and have come to appreciate its performant well crafted design.

In case anyone else runs into the issue...

In LV 2025 the ability to open reentrant clones on the cRIO is supported in the LV IDE. I have confirmed that I can open the front panel and probe the BD of reentrant clones running on my cRIO with Linux RT. There is a bug however that inhibits this ability until the execution properties window of the reentrant vi is opened and the code restarted.

Experiencing strange behavior on a cRIO where I have a heater controller that uses 3 PID vi's. The execution properties of the PID vi's are set to Preallocated clone Reentrant. When running from the project the individual PID vi's do not initially open as clones. Only after opening the properties and then restarting do things start working properly. This is a repeatable issue that reoccurs each time the code is deployed to the cRIO from the project. I suppose I could just save the PID vi multiple times with different name but that is a poor solution and cumbersome since the PID vi is in the NI_PID_autopid.lvlib.

Just to wanted add another voice to the LuaVIEW fan club. LUA is an excellent scripting language and LuaVIEW is an excellent integration with LV. Keep hoping you might see the advantages of open sourcing it to the LV community.

Noticed that MGI RWA does not seem to work when clusters contain things like IMAQdx refnums. Seems like the default Major Vers will always be 1 which is incorrect and causes conversion back to variant problems downstream. When I changed the default of the Scan String first parameter to 2 then it works fine. This is my cluster that I am trying to work with. The problem occurs with the IMAQdx refnum control.

Bridgeview? You must be old as the hills I think if back then they had our modern DDS options We would have never had to suffer NSV’s and the product might be still viable today.

Remember the venerable DSC toolkit? For its time it was a well thought out collection of SCADA tools, historical database, distributed systems manager, Network tag engine, etc. Now that we are NI/Emerson can we hope for a modern version of this? Sans Network Shared Variables of course! I suppose there is Systemlink but I am not sure that really covers SCADA usecases.