search

prometheus-community / helm-charts

Prometheus community Helm charts
Apache License 2.0
4.98k stars 4.99k forks source link

[kube-prometheus-stack] Networking NO DATA #4788

Open jsalatiel opened 1 month ago

jsalatiel commented 1 month ago

Describe the bug a clear and concise description of what the bug is.

I have installed prometheus-kube-stack with the default values.

So , my values is just:

grafana:
  enabled: true
  ingress:
    enabled: true
    hosts:
      - k8s.my.domain

When I login to grafana, all default dashboards related to network just show NO DATA.

What's your helm version?

v3.15.2

What's your kubectl version?

v1.29.2

Which chart?

kube-prometheus-stack

What's the chart version?

61.9.0

What happened?

NO DATA for all network related dashboards

What you expected to happen?

I expect the network information be populated in the default dashboards.

How to reproduce it?

Install the helm chart with default values on a single node k8s .

Enter the changed values of values.yaml?

grafana:
  enabled: true
  ingress:
    enabled: true
    hosts:
      - k8s.my.domain

Enter the command that you execute and failing/misfunctioning.

helm upgrade -i prometheus prometheus/kube-prometheus-stack -n monitoring -f myvalues.yml

Anything else we need to know?

I run antrea 2.1.0 as CNI

jsalatiel commented 1 month ago

image image

kyrbrbik commented 1 month ago

I have the same issue on some of our installations. All are running on 61.9.0 version of the chart. I suspect that there is a problem with this part of the query * on (cluster,namespace,pod) kube_pod_info{host_network="false"} for some reason. They all error out with execution: multiple matches for labels: many-to-one matching must be explicit (group_left/group_right)

jsalatiel commented 4 weeks ago

maybe related https://github.com/google/cadvisor/issues/3577

sba-ops commented 3 weeks ago

I'm experiencing the same issues with the helm chart versions 61.2.0 and 62.2.1.

rahuks123 commented 2 weeks ago

i too am experiencing a similar issue where any dashboard related to network is showing no-data i have attached the screenshot of the page i am getting. i checked the helm chart values.yaml but if has cadvisor enabled which i assume should populate these dashboards image

jsalatiel commented 2 weeks ago

i too am experiencing a similar issue where any dashboard related to network is showing no-data i have attached the screenshot of the page i am getting. i checked the helm chart values.yaml but if has cadvisor enabled which i assume should populate these dashboards image

Cri-o?

rahuks123 commented 2 weeks ago

i am using a cloud provisioned Kubernetes where i have directly applied the helm chart of kube-prometheus

jsalatiel commented 2 weeks ago

i am using a cloud provisioned Kubernetes where i have directly applied the helm chart of kube-prometheus

What CRI does your cloud provider uses for its backend?

rahuks123 commented 2 weeks ago

dont have much idea. i use IBMcloud provisioned kubernetes. if i search in net i get CRI saying 'containerd', but not sure if that is what u asked...

jseiser commented 2 weeks ago

I believe its because the variable cluster is returning no information.

rahuks123 commented 5 days ago

i tried another release as they had mentioned tag which deals with network data failing. installed using :

helm install kube-prometheus-stack prometheus-community/kube-prometheus-stack --version 62.5.1 -n monitoring and now i am able to see the network data for my namespaces