I am trying to figure out what the best way to run Grafana/Heapster/InfluxDB on my GKE Kubernetes cluster.
From what I understand I cannot modify the kube-system
namespace unless I want google to overwrite these settings after a time. So changing the heapster that is running in kube-system is probably not an option.
Does that mean I need another heapster in my default namespace alongside grafana and InfluxDB pods?