This message was deleted.

@dat Can you clarify what use cases you are trying to achieve using these tools?

At a high level, observability. Grafana will be our UI for metrics, logging, tracing of services in a given k8s cluster. The dashboard in question would be used for general monitoring of cluster resources. I know alerting is something we’re going to implement, e.g. when memory usage is above a certain threshold. At present, however, this grafana dashboard, 315, is reporting only on cluster level metrics and no data for Pod metrics like cpu, memory, network I/O.

I am a bit biased but I recommend you to check out www.perfectscale.io as they provide most of the data you are looking for and do much more than just observability.

Thanks for the rec, will def take a look

Did you check this out? https://monitoring.mixins.dev/kubernetes/#dashboards

If you got the money datadog is pretty great too

You could also look at https://pixielabs.ai/ as another CNCF project for cluster observability. If you're a small team or you don't see a high volume of events, New Relic has a free tier with out of the box alerts (100GB of data/mo, 1 free full platform user, unlimited alerts/dashboards, etc). Disclosure on the latter: Former New Relic employee here.

Grafana do actually have a Kubernetes integration now with some OOTB dashboards: https://grafana.com/blog/2022/07/13/introducing-kubernetes-monitoring-in-grafana-cloud/, super useful and it's all Prometheus based!