to limit the number of reschedules as we cannot guarantee an error-free job definition.
We also set reasonable delays and intervals to increase the likelihood that one deployment has enough time, but there is no second deployment within the interval.