Applying persistent host-group settings to ROSA worker nodes

[Greg_Marwick]

We're running a ROSA (Red Hat OpenShift Service on AWS) environment where worker nodes are frequently spun up and down. We need two specific settings applied to these nodes:

1. Disk alert suppression – A commonly used Linux directory on these hosts is intentionally kept at capacity by design, causing false-positive low disk alerts due to our global alert settings.
2. Graceful shutdown alert suppression – Since these nodes are expected to be regularly terminated and recreated, we want to ignore graceful shutdown alerts for them specifically.

The problem is that any host-level settings we apply are lost when a node is recycled, the best solution I can think of is to assign these a host-group when they are created, which would allow the settings to persist. However, I'm unsure how to configure the Dynatrace Operator to automatically assign all worker nodes in this ROSA environment to a host group when they are created.

Is there a way to define a host group for all worker nodes via the Dynatrace Operator in a ROSA environment, so that the correct settings are consistently applied regardless of node churn?

[dannemca]

Have you tried setting the following in your dynakube?

...
spec:
  apiUrl: https://ENVIRONMENTID.live.dynatrace.com/api
  oneAgent:
    # Optional: Set a host group for OneAgent
    #
    hostGroup: "MyHostGroup"
...

Regards.