Deploying on the Kwok node, unable to collect the kube_pod_datus_stcheduled_time metric

[conghuhu]

How to use it?

• kwok
• kwokctl --runtime=docker (default runtime)
• kwokctl --runtime=binary
• kwokctl --runtime=nerdctl
• kwokctl --runtime=kind

What happened?

The pod deployed on the Kwok node cannot collect the kube_pod_datus_stcheduled_time metric in kube state metrics.``

What did you expect to happen?

should query result by count(kube_pod_status_scheduled_time{namespace="default"}) in prometheus

How can we reproduce it (as minimally and precisely as possible)?

1. use follow shell to install kube-prometheus-stack

#!/bin/bash
helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
helm repo update

helm install --set nodeExporter.enabled=false,alertmanager.enabled=false,prometheusOperator.admissionWebhooks.patch.image.registry="docker.io",prometheusOperator.admissionWebhooks.patch.image.repository=15841721425/kube-webhook-certgen kube-prometheus-stack --set kube-state-metrics.image.registry="docker.io",kube-state-metrics.image.repository=bitnami/kube-state-metrics,kube-state-metrics.image.tag=2.10.1 prometheus-community/kube-prometheus-stack -n monitoring --create-namespace

2. use follow shell to install kwok

#!/bin/bash

if [ $# -eq 0 ]; then
  echo "Error: Please provide the number of nodes to create."
  echo "Usage: $0 <number_of_nodes>"
  exit 1
fi

KWOK_REPO=kubernetes-sigs/kwok
KWOK_LATEST_RELEASE=$(curl "https://api.github.com/repos/${KWOK_REPO}/releases/latest" | jq -r '.tag_name')
kubectl apply -f "https://github.com/${KWOK_REPO}/releases/download/${KWOK_LATEST_RELEASE}/kwok.yaml"
kubectl apply -f "https://github.com/${KWOK_REPO}/releases/download/${KWOK_LATEST_RELEASE}/stage-fast.yaml"

for (( i=0;i<$1; i++))
do
  kubectl apply -f - <<EOF
  apiVersion: v1
  kind: Node
  metadata:
    annotations:
      node.alpha.kubernetes.io/ttl: "0"
      kwok.x-k8s.io/node: fake
    labels:
      beta.kubernetes.io/arch: amd64
      beta.kubernetes.io/os: linux
      kubernetes.io/arch: amd64
      kubernetes.io/hostname: kwok-node-$i
      kubernetes.io/os: linux
      kubernetes.io/role: agent
      node-role.kubernetes.io/agent: ""
      type: kwok
    name: kwok-node-$i
  spec:
    taints: # Avoid scheduling actual running pods to fake Node
    - effect: NoSchedule
      key: kwok.x-k8s.io/node
      value: fake
  status:
    allocatable:
      cpu: 32
      memory: 256Gi
      pods: 110
    capacity:
      cpu: 32
      memory: 256Gi
      pods: 110
    nodeInfo:
      architecture: amd64
      bootID: ""
      containerRuntimeVersion: ""
      kernelVersion: ""
      kubeProxyVersion: fake
      kubeletVersion: fake
      machineID: ""
      operatingSystem: linux
      osImage: ""
      systemUUID: ""
    phase: Running
EOF
done

3. deploy a nginx deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  replicas: 2
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
        - name: nginx
          image: nginx
          ports:
            - containerPort: 80
      tolerations:
        - key: "kwok.x-k8s.io/node"
          operator: "Exists"
          effect: "NoSchedule"
---
apiVersion: v1
kind: Service
metadata:
  name: nginx-service
spec:
  type: NodePort
  ports:
    - port: 80
      targetPort: 80
  selector:
    app: nginx

4. query count(kube_pod_status_scheduled_time{namespace="default"}) in prometheus

Anything else we need to know?

No response

Kwok version

$ kwok --version
0.6.0

$ kwokctl --version
# paste output here

OS version

```console # On Linux: $ cat /etc/os-release # paste output here $ uname -a # paste output here

On Darwin:

$ uname -a

paste output here

On Windows:

C:> wmic os get Caption, Version, BuildNumber, OSArchitecture

paste output here

</details>

[wzshiming]

https://kwok.sigs.k8s.io/docs/user/metrics-configuration/

These are configurable, and you must configure the metrics as you see fit.

/remove-kind bug
/kind support

[conghuhu]

https://kwok.sigs.k8s.io/docs/user/metrics-configuration/

These are configurable, and you must configure the metrics as you see fit.

/remove-kind bug /kind support

OKay, I will try

[conghuhu]

@wzshiming May I ask if this metric is simulated or real based on document? I would like to expose the real metric

[wzshiming]

There are only simulated metrics. what exactly do you mean by real metrics?

[conghuhu]

There are only simulated metrics. what exactly do you mean by real metrics?

I want to track the time it takes for pods on the Kwok node to reach the scheduled state. Currently, I am using the kube_pod_datus_stcheduled_time metric provided by kube-state-metrics, but I am unable to collect metric data on the Kwok node

[wzshiming]

you also need to use kube-state-metrics to provide the metrics you need

[conghuhu]

kube-state-metrics can already be collected. I deployed nginx on a non kwok node, which can collect metrics, but not on a kwok node

[wzshiming]

Ahh, it seems that kwok needs to be adapted to kube-state-metrics

[conghuhu]

Excuse me, how long will it take to adapt?

[wzshiming]

This requires kube-state-metrics to adapt to kwok and it to support collecting multiple nodes simulated by kwok.

https://github.com/kubernetes/kube-state-metrics/blob/f8aa7d9bb9d8e29876e19f4859391a54a7e61d63/pkg/app/server.go#L228-L233