---
title: "Comment by Damiano Bianda on \"Stop Pinging Kafka - A Zero-Cost Health Check Pattern (and What Can Go Wrong)\" | Hashnode"
description: "The \"the health check itself made the cluster less reliable\" pattern shows up far beyond Kafka, and your GC-pauses-to-restart-storm chain is the sharpest exampl"
url: "https://hashnode.com/posts/stop-pinging-kafka-a-zero-cost-health-check-pattern-and-what-can-go-wrong/6ad65e65cb89e7ce1f6ae2c7/comment/6ad71841044eb377c226fae7"
---

[Discussion](https://hashnode.com/posts/stop-pinging-kafka-a-zero-cost-health-check-pattern-and-what-can-go-wrong/6ad65e65cb89e7ce1f6ae2c7)

Comment [Stop Pinging Kafka - A Zero-Cost Health Check Pattern (and What Can Go Wrong)](https://hashnode.com/posts/stop-pinging-kafka-a-zero-cost-health-check-pattern-and-what-can-go-wrong/6ad65e65cb89e7ce1f6ae2c7)

[Damiano Bianda](https://hashnode.com/@damianobianda)

[Oct 20](https://hashnode.com/posts/stop-pinging-kafka-a-zero-cost-health-check-pattern-and-what-can-go-wrong/6ad65e65cb89e7ce1f6ae2c7/comment/6ad71841044eb377c226fae7)

The "the health check itself made the cluster less reliable" pattern shows up far beyond Kafka, and your GC-pauses-to-restart-storm chain is the sharpest example of the monitor and the monitored sharing a fate. The failure mode I'd want a follow-up on: how do you notice when the piggybacked statistics themselves stop arriving? If the callback stalls silently, you end up with the calmest dashboard of all - the one that says nothing is wrong because nothing is being measured. Did you end up putting a watchdog on the absence of callbacks, or is staleness just accepted as a blind spot?