Exercise 2: Resource Management

Task 1 - Understanding Resource Requests and Limits

Properly setting resource requests and limits is essential for optimal cluster performance. Let’s see how to optimize resource usage:

  1. Create a namespace for a QoS (Quality of Service) demonstration:

    ​
    kubectl create namespace qos-demo
    kubectl create namespace qos-demo

  2. Create a Guaranteed QoS pod (equal requests and limits):

    ​
    $guaranteedPodYaml = @"
apiVersion: v1
kind: Pod
metadata:
  name: guaranteed-pod
  namespace: qos-demo
spec:
  containers:
  - name: nginx
    image: k8sonazureworkshoppublic.azurecr.io/nginx
    resources:
      requests:
        memory: "100Mi"
        cpu: "100m"
      limits:
        memory: "100Mi"
        cpu: "100m"
"@

$guaranteedPodYaml | kubectl apply -f -
    cat << EOF | kubectl apply -f -
apiVersion: v1
kind: Pod
metadata:
  name: guaranteed-pod
  namespace: qos-demo
spec:
  containers:
  - name: nginx
    image: k8sonazureworkshoppublic.azurecr.io/nginx
    resources:
      requests:
        memory: "100Mi"
        cpu: "100m"
      limits:
        memory: "100Mi"
        cpu: "100m"
EOF

  3. Create a Burstable QoS pod (requests less than limits):

    ​
    $burstablePodYaml = @"
apiVersion: v1
kind: Pod
metadata:
  name: burstable-pod
  namespace: qos-demo
spec:
  containers:
  - name: nginx
    image: k8sonazureworkshoppublic.azurecr.io/nginx
    resources:
      requests:
        memory: "50Mi"
        cpu: "50m"
      limits:
        memory: "100Mi"
        cpu: "100m"
"@

$burstablePodYaml | kubectl apply -f -
    cat << EOF | kubectl apply -f -
apiVersion: v1
kind: Pod
metadata:
  name: burstable-pod
  namespace: qos-demo
spec:
  containers:
  - name: nginx
    image: k8sonazureworkshoppublic.azurecr.io/nginx
    resources:
      requests:
        memory: "50Mi"
        cpu: "50m"
      limits:
        memory: "100Mi"
        cpu: "100m"
EOF

  4. Create a BestEffort QoS pod (no requests or limits):

    ​
    $bestEffortPodYaml = @"
apiVersion: v1
kind: Pod
metadata:
  name: besteffort-pod
  namespace: qos-demo
spec:
  containers:
  - name: nginx
    image: k8sonazureworkshoppublic.azurecr.io/nginx
"@

$bestEffortPodYaml | kubectl apply -f -
    cat << EOF | kubectl apply -f -
apiVersion: v1
kind: Pod
metadata:
  name: besteffort-pod
  namespace: qos-demo
spec:
  containers:
  - name: nginx
    image: k8sonazureworkshoppublic.azurecr.io/nginx
EOF

  5. Check the QoS class of each pod:

    ​
    kubectl get pods -n qos-demo -o custom-columns="NAME:.metadata.name,QOS:.status.qosClass"
    kubectl get pods -n qos-demo -o custom-columns="NAME:.metadata.name,QOS:.status.qosClass"

    You should see Guaranteed, Burstable, and BestEffort respectively.

  6. When the cluster is under resource pressure, Kubernetes will evict pods in this order: BestEffort, Burstable, and finally Guaranteed. This is important to understand when planning workload priorities.

Task 2 - Monitor Resource Usage

  1. Check current resource usage for your pods:

    ​
    kubectl top pods -n qos-demo
    kubectl top pods -n qos-demo
    Tip

    If you receive a message stating error: metrics not available yet, you may need to wait a few moments for metrics to be collected.

    The top command shows the actual resources in use by each pod, rather than the requested or limited resources.

  2. Check node resource usage:

    ​
    kubectl top nodes
    kubectl top nodes

Task 3 - Create Namespace with Resource Quotas

  1. Create a new namespace for testing resource quotas:

    ​
    kubectl create namespace limited-resources
    kubectl create namespace limited-resources

  2. Create a ResourceQuota in this namespace:

    ​
    $quota = @"
apiVersion: v1
kind: ResourceQuota
metadata:
  name: mem-cpu-quota
  namespace: limited-resources
spec:
  hard:
    requests.cpu: "1"
    requests.memory: 1Gi
    limits.cpu: "2"
    limits.memory: 2Gi
    pods: "5"
"@

$quota | kubectl apply -f -
    cat << EOF | kubectl apply -f -
apiVersion: v1
kind: ResourceQuota
metadata:
  name: mem-cpu-quota
  namespace: limited-resources
spec:
  hard:
    requests.cpu: "1"
    requests.memory: 1Gi
    limits.cpu: "2"
    limits.memory: 2Gi
    pods: "5"
EOF

  3. Verify the ResourceQuota:

    ​
    kubectl describe resourcequota -n limited-resources
    kubectl describe resourcequota -n limited-resources

Task 4 - Test Resource Quota Enforcement

  1. Create a deployment that uses a lot of the memory quota:

    ​
    $deployment1 = @"
apiVersion: apps/v1
kind: Deployment
metadata:
  name: memory-demo
  namespace: limited-resources
spec:
  replicas: 1
  selector:
    matchLabels:
      app: memory-demo
  template:
    metadata:
      labels:
        app: memory-demo
    spec:
      containers:
      - name: memory-demo
        image: k8sonazureworkshoppublic.azurecr.io/nginx
        resources:
          requests:
            memory: "800Mi"
            cpu: "500m"
          limits:
            memory: "1.5Gi"
            cpu: "1"
"@

$deployment1 | kubectl apply -f -
    cat << EOF | kubectl apply -f -
apiVersion: apps/v1
kind: Deployment
metadata:
  name: memory-demo
  namespace: limited-resources
spec:
  replicas: 1
  selector:
    matchLabels:
      app: memory-demo
  template:
    metadata:
      labels:
        app: memory-demo
    spec:
      containers:
      - name: memory-demo
        image: k8sonazureworkshoppublic.azurecr.io/nginx
        resources:
          requests:
            memory: "800Mi"
            cpu: "500m"
          limits:
            memory: "1.5Gi"
            cpu: "1"
EOF

  2. Try to create a second deployment that would exceed the quota:

    ​
    $deployment2 = @"
apiVersion: apps/v1
kind: Deployment
metadata:
  name: memory-demo2
  namespace: limited-resources
spec:
  replicas: 1
  selector:
    matchLabels:
      app: memory-demo2
  template:
    metadata:
      labels:
        app: memory-demo2
    spec:
      containers:
      - name: memory-demo2
        image: k8sonazureworkshoppublic.azurecr.io/nginx
        resources:
          requests:
            memory: "800Mi"
            cpu: "500m"
          limits:
            memory: "1.5Gi"
            cpu: "1"
"@

$deployment2 | kubectl apply -f -
    cat << EOF | kubectl apply -f -
apiVersion: apps/v1
kind: Deployment
metadata:
  name: memory-demo2
  namespace: limited-resources
spec:
  replicas: 1
  selector:
    matchLabels:
      app: memory-demo2
  template:
    metadata:
      labels:
        app: memory-demo2
    spec:
      containers:
      - name: memory-demo2
        image: k8sonazureworkshoppublic.azurecr.io/nginx
        resources:
          requests:
            memory: "800Mi"
            cpu: "500m"
          limits:
            memory: "1.5Gi"
            cpu: "1"
EOF

  3. Observe the error message indicating the quota has been exceeded by reviewing the replicaset:

    ​
    kubectl describe rs -l "app=memory-demo2" -n limited-resources
    kubectl describe rs -l "app=memory-demo2" -n limited-resources

  4. Check the current quota usage:

    ​
    kubectl describe resourcequota -n limited-resources
    kubectl describe resourcequota -n limited-resources

Task 5 - LimitRange for Default Values

  1. Create a LimitRange to set default resource limits and requests:

    ​
    $limitRange = @"
apiVersion: v1
kind: LimitRange
metadata:
  name: default-limits
  namespace: limited-resources
spec:
  limits:
  - default:
      cpu: "500m"
      memory: "256Mi"
    defaultRequest:
      cpu: "100m"
      memory: "128Mi"
    type: Container
"@

$limitRange | kubectl apply -f -
    cat << EOF | kubectl apply -f -
apiVersion: v1
kind: LimitRange
metadata:
  name: default-limits
  namespace: limited-resources
spec:
  limits:
  - default:
      cpu: "500m"
      memory: "256Mi"
    defaultRequest:
      cpu: "100m"
      memory: "128Mi"
    type: Container
EOF

  2. Create a pod without specifying resources:

    ​
    $defaultPod = @"
apiVersion: v1
kind: Pod
metadata:
  name: default-resources-pod
  namespace: limited-resources
spec:
  containers:
  - name: default-resources
    image: k8sonazureworkshoppublic.azurecr.io/nginx
"@

$defaultPod | kubectl apply -f -
    cat << EOF | kubectl apply -f -
apiVersion: v1
kind: Pod
metadata:
  name: default-resources-pod
  namespace: limited-resources
spec:
  containers:
  - name: default-resources
    image: k8sonazureworkshoppublic.azurecr.io/nginx
EOF

  3. Check that default resources were applied:

    ​
    kubectl get pod default-resources-pod -n limited-resources -o jsonpath='{.spec.containers[0].resources}' | ConvertFrom-Json
    kubectl get pod default-resources-pod -n limited-resources \
  -o jsonpath='Requests: cpu={.spec.containers[0].resources.requests.cpu}, memory={.spec.containers[0].resources.requests.memory}{"\n"}Limits:   cpu={.spec.containers[0].resources.limits.cpu},   memory={.spec.containers[0].resources.limits.memory}{"\n"}'

    Look for the “Requests” and “Limits” sections to confirm the LimitRange defaults were applied.

Task 6 - Cleanup

  1. Delete the namespace and all resources created in this exercise:

    ​
    kubectl delete namespace limited-resources
kubectl delete resourcequota mem-cpu-quota -n limited-resources
kubectl delete limitrange default-limits -n limited-resources
kubectl delete pod default-resources-pod -n limited-resources
kubectl delete ns qos-demo
    kubectl delete namespace limited-resources
kubectl delete resourcequota mem-cpu-quota -n limited-resources
kubectl delete limitrange default-limits -n limited-resources
kubectl delete pod default-resources-pod -n limited-resources
kubectl delete ns qos-demo