Reece, do shortcuts commonly select different files, even with the same script?

They can. Compare the actual working directory and loaded configuration rather than assuming the nearby file is the one used

Then preserve the current failing setup before the paired comparison. The working result is useful evidence of one success, but it does not yet establish that the endpoint, route or software environment matches now.

Who owns that comparison, Grace? Controller owner for intended target, network owner for permitted route and service, maintainer for the saved launch. Keep their answers together.

Our maintainer is coordinating all three. The failing shortcut loads a personal training profile, not the documented cell profile.

Does that profile actually name a different destination? I'd write down the loaded values before calling the profile the cause. A wrong-looking file name can still contain the right address, which would leave the timeout unexplained.

Different destination confirmed. Network owner verified the intended cell route; maintainer is correcting the profile selection, not changing controller settings.

That answers my shortcut question too, thank you Lin. The values mattered, not the file's reassuring name.

Has the intended read succeeded after that correction, Grace? I would keep target mismatch found and service check passed as two results until the second actually happens.

And repeat from the ordinary shortcut, not only the maintainer's terminal.

Sarah's launch point belongs in the rebuild instructions too. If the fix depends on knowing which menu choice not to use, the next shift may reproduce the failure quite faithfully.

Normal shortcut now returns the intended FR5 identity. Reopening also passes. Both sessions retain target, route and SDK details.

Keep the old profile identified as training-only or remove it from the operational choices under your configuration process. It remains useful evidence of the fault, not a second equally valid launch option.

Operational shortcut fixed and training profile separated. Maintainer updated the rebuild note. This timeout is closed without a controller change.

Give the covering maintainer that note too. The paired evidence is more useful now than the earlier claim that another laptop worked.