longhorn

longhorn copied to clipboard

Describe the bug

I've been configuring a monitoring for longhorn v1.3.0 according to the docs https://longhorn.io/docs/1.3.0/monitoring/prometheus-and-grafana-setup/

In the end I've imported grafana dashboard provided by the docs. Some of the panels related to CPU / Memory usage show 'No Data'

I've looked into the metrics and it comes manger doesn't provide these longhorn_node_cpu_usage_millicpu (available in docs), while the longhorn_node_cpu_capacity_millicpu metrics are provided. This is also related to memory usage as well

• minor improvement consideration: latest releases of kube-prometheus-stack uses label release: <release name> to scrap the metrics. Might be good to mention this in docs in conjunction with namespaces tips to simplify configuration for users Many thanks

Environment

• Longhorn version: 1.3.0
• Installation method (e.g. Rancher Catalog App/Helm/Kubectl): helm
• Kubernetes distro (e.g. RKE/K3s/EKS/OpenShift) and version: 1.24.3
  • Number of management node in the cluster: 1
  • Number of worker node in the cluster: 2
• Node config
  • OS type and version: Ubuntu 20.04
  • CPU per node: 5
  • Memory per node: 28
  • Disk type(e.g. SSD/NVMe): SSD
  • Network bandwidth between the nodes:
• Underlying Infrastructure (e.g. on AWS/GCE, EKS/GKE, VMWare/KVM, Baremetal): Baremetal
• Number of Longhorn volumes in the cluster: 2

Additional context

Add any other context about the problem here.

cc @c3y1huang