I used the concepts described in the example code posted with the article titled '
Access TestStand Variables and Properties from an Executable Code Module' (http://sine.ni.com/apps/utf8/niepd_web_display.display_epd4?p_guid=B45EACE3D9EE56A4E034080020E74861), and developed a CVI executable that accesses teststand variables in the context of the sequence that calls it (let's call this sequence B). Everything works fine there. This piece is a small component of a broader distributed application, therefore I need to be able to execute sequence B on computer B from sequence A on computer A, using the remote execution option for the sequence call step in sequence A. I can successfully use remote execution between computer A and B with other sequences, so I know DCOM is setup properly. However, calling sequence B on computer B from sequence A on computer A doesn't yield the expected result. Suspecting my implementation to be the culprit, I tested the same concept with the example code, unmodified, from the example mentioned above. That test consists of a test sequence A on computer A calling the example sequence GetSeqContextOrEngineHandleFromEXE.seq on remote computer B. This test generates an error during execution of the step 'Display Locals-Num' because the executable ExecutableWithSeqContextHandle.exe doesn't successfully create Locals.Num in the sequence context of GetSeqContextOrEngineHandleFromEXE.seq. Again if this example is executed by itself it works fine.
I need to understand how the remote execution affects the use of the teststand API from an executable, and how it affects the registration of the sequence context for use by the EXE (I have a feeling that might be the issue). And I also need to know how to make this work.
Any thoughts on this would be greatly appreciated. Thx!