Varnish operator grafana dashboard pulls in too many metrics
kscharm opened this issue · 2 comments
kscharm commented
The Grafana dashboard that is deployed with the operator pulls in metrics from other pods in the cluster that are unrelated to varnish. See the CPU usage and Memory usage graphs in the dashboard. We should narrow the scope of the metrics that are used in the dashboard. Additionally, the CPU usage graph appears to use time when it should be using percentage.
kscharm commented
@tomashibm let me know if you have anything else to add here ^
tomashibm commented
I'd also set a filter by the namespace. Wouldn't hurt and if we decide to support deploying multiple operators in a cluster (but in different namespaces), this part will be ready for that feature.