the customer has an OpenShift cluster, which operates on two different segments: PSI and PROM.
We use different Dynatrace clusters to monitor PSI and Prom.
1. How to properly install and configure OneAgent Operator so that we can monitor PSI and PROM in different Dynatrace installations, while OneAgent Operator does not interfere with each other and data on containers from PROM does not get into the PSI stand (and vice versa)
2. How to correctly take metrics using the Kebernets API so that data on nodes and submissions from PROM does not get into the PSI stand (and vice versa)
Solved! Go to Solution.
Our new injection mechanism, using the Kubernetes admission controller, will be available for full stack monitoring later in 2021. For now, you can use it for application only monitoring. By deploying two custom resources, with different annotations, you can direct metrics to do different environments.
Again, this will not work in a full stack mode for at least a few months, but we're getting there.