I have a Kubernetes cluster with nodes running Talos Linux. It auto-detects the Prometheus operator running on my cluster, but only the memory metrics are available. CPU & Disk metrics are missing.
Hello clarknova9,
Thank you for reaching out to Lens support!
Could you tell us how the Prometheus was installed, please? With helm or something else?
We are waiting for you to reply.
Best Regards,
Tikhon Kudinov
Lens Support Engineer
Mirantis, Inc
Hi Tikhon. I have a kustomization file with the HelmRelease configuration. I’m using the kube-prometheus-stack chart. I do see it the Prometheus operator getting auto-detected by Lens when I go into the seetings:
AUTO DETECTED PROMETHEUS DETAILS
Provider: Prometheus Operator
Path: observability/services/prometheus-operated:9090
I’ve ran the same config on a k3s cluster, and Lens did not have this issue.
Hello clarknova9,
Could you try to set the Helm or Helm 14.x and try again, please?
We are waiting for you to reply.
Best Regards,
Tikhon Kudinov
Lens Support Engineer
Mirantis, Inc
Switching to Helm 14.x appears to work better, but it also has issues. Here is what is working / not working when I choose different options.
Autodetect
- Cluster View: Memory & Pod usage is available. CPU usage is missing.
- Nodes View: Memory usage available. CPU & disk usage is missing.
- Pod View: Memory usage is available. CPU, Network & disk usage is missing.
Helm 14.x
- Cluster View: CPU and memory usage is available. Pods usage is missing.
- Nodes View: CPU usage is available. Disk and memory is missing.
- Pod View: CPU, memory and network usage is available. Disk usage is missing.
Helm
- Cluster View: All metrics missing.
- Nodes view: All metrics missing.
- Pod View: All metrics missing.
Prometheus Operator
- Cluster View: Memory & Pod usage is available. CPU usage is missing.
- Nodes View: Disk usage is available. CPU & Memory usage is missing.
- Pod View: Memory usage is available. CPU, Network & disk usage is missing.
Hello clarknova9,
Do you have metrics-server running in your cluster? If not, install it, please.
We are waiting for you to reply.
Best Regards,
Tikhon Kudinov
Lens Support Engineer
Mirantis, Inc
I have metrics-server running. It’s the metrics server installed from this chart: Kubernetes Metrics Server | metrics-server
Hello clarknova9,
Thank you for the answer.
We have one more question.
What version of that chart are you using?
We are waiting for you to reply.
Best Regards,
Tikhon Kudinov
Lens Support Engineer
Mirantis, Inc
Hello clarknova9,
Also, we have some kind of feature request.
You can check it too.
Configurable mountpoint or custom metric query for Node disk avail / used? · Issue #7979 · lensapp/lens · GitHub
Best Regards,
Tikhon Kudinov
Lens Support Engineer
Mirantis, Inc
I am running the latest version of both metrics-server and kube-prometheus-stack.
metrics-server: 3.12.1
kube-prometheus-stack: 61.5.0