But we noticed an interesting point. When the user reconnected to the test Fleet, the watchdog disable status recognition began to occur correctly. Moreover, it started to happen correctly after the user returned to the main Fleet. That is, it seems that some kind of data refresh was required. At the same time, I can say that other settings, for example, the amount of RAM allocated for the watchdog (watchdog_memory_limit), were recognized correctly even when they were changed. Therefore, we want to re-initialize the Fleet database and see if this helps in solving our problem.