Correct, Ideally monitoring stack should be outside the blast radius of other applications. Will handling another Kafka cluster (probably smaller than business Kafka) be a pain for the team given the team already knows managing one business Kafka. What do you think?