Controller owner confirmed the documented endpoint. The launchers import the same supported SDK and issue the same read, but the failing laptop resolves the configured host name to another address.

Is that wrong address also the destination of its successful pings?

Alex's question would explain why reachability looked encouraging, but don't assume it yet. Record the destinations. An answering device is a useful observation only when you know which device answered.

Yes, the successful pings used that other address. Network owner found an obsolete local name entry on the failing laptop; the working laptop resolves the same name to the documented controller address.

Have the responsible owner correct the obsolete entry under the approved configuration, then repeat the documented service read from the normal launcher. Resolution and route evidence narrow the cause, but successful ping after correction would not complete that check.

And keep the original resolved address in the repair note. Otherwise the next person sees a correct host name in both configurations and can't understand what was different. The text in the settings box was not the whole endpoint.

The owner corrected the entry. Our normal launcher now completes the documented read against the intended controller, with its selected route recorded. Reopening the application gives the same result; we retained the before-and-after endpoint evidence.

That's a concrete answer to the laptop difference. No need to dress it as an unreliable controller service when the failing request had been going elsewhere. Have the rebuild notes stopped reproducing that old entry too?

Give the maintainer the verified launcher, endpoint-resolution requirement and network-owner reference together. The service test establishes this corrected path; it does not certify every future network configuration merely because the same host name appears in it.

Thanks for tracing the ping destination. That was the misleading success worth explaining.