If someone changes a cluster option that sets a controller timer duration (dc-deadtime, join-integration-timeout, etc.), config_query_callback() -> controld_configure_fsa_timers() will update it.

Currently, this does not affect any existing timer that's already active, which will continue with the old value. We should probably restart the timer to make the new value take effect. (Even better would be to add the difference between the old and new values to the remaining timeout of the current instance, but that's probably impractical.)