Install the Cluster Agent with the Kubernetes CLI

This page describes how to install the Cluster Agent using the Kubernetes CLI which is an alternative toInstall the Cluster Agent with Helm Charts.

Requirements

Before you begin, verify that you have:

Installation Procedure

  1. If you have not already installed the metrics-server https://github.com/kubernetes-sigs/metrics-server for the most recent installation instructions.
    kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml
  2. Download the Kubernetes or OpenShift Cluster Agent bundle from the Download portal, unzip the contents of the bundle to the current directory:
    unzip appdynamics-cluster-agent-alpine-linux-<version>.zip
  3. Create a namespace appdynamics in the cluster. If you require to use a custom namespace, you can ignore this step and ensure to use the same namespace during installation. Here, we have used namespace as appdynamics.
    kubectl create namespace appdynamics
  4. (Optional) If you require to use the Kubernetes API server, ensure that you set the USE_CUSTOM_KUBECONFIG env variable to true and specify the server API details in the cluster-agent-operator.yaml file. For details, see Configure Image to Use Kubernetes API Server.
  5. Install the Cluster Agent Operator using the correct Kubernetes and OpenShift version (where applicable):
    Kubernetes >= 1.16
    kubectl create -f cluster-agent-operator.yaml
    OpenShift with Kubernetes >= 1.16
    kubectl create -f cluster-agent-operator-openshift.yaml
    OpenShift with Kubernetes <= 1.15
    kubectl create -f cluster-agent-operator-openshift-1.15-or-less.yaml
    Note: You can also install Cluster Agent Operator from OpenShift OperatorHub in your OpenShift cluster.
  6. Based on the Account Access Key for the Controller, create the Controller Access Key Secret that the Cluster Agent reports to:
    kubectl -n appdynamics create secret generic cluster-agent-secret --from-literal=controller-key=<access-key>
  7. Edit cluster-agent.yaml nsToMonitorRegexThis example assumes the use of the pre-built Cluster Agent image on Docker Hub. See Configure the Cluster Agent, Cluster Agent Container Image to build your own image.
    apiVersion: cluster.appdynamics.com/v1alpha1
kind: Clusteragent
metadata:
  name: k8s-cluster-agent
  namespace: appdynamics
spec:
  appName: "dev-cluster"
  controllerUrl: "http://mycontroller.com:8080"
  account: "my-account"
  # docker image info where, <version> is the image tag version that you are installing
  image: "docker.io/appdynamics/cluster-agent:<version>"
  serviceAccountName: appdynamics-cluster-agent
  nsToMonitorRegex: namespace1|namespace2
  stdoutLogging: "true"
  8. (Optional) If you require multiple Cluster Agents to monitor a single cluster, set up Target Allocator. In thecluster-agent.yaml file, enable target allocator with the number of Cluster Agent replicas to enable the operator to create Cluster Agent replicas. See Target Allocator.
    Note: Auto-instrumentation and containerAppCorrelationMethod (for Kubernetes >= 1.25) is not supported when the Target Allocator is enabled. Set spec.instrumentationMethod to "" or "None" and spec.containerAppCorrelationMethod to ""or "none".

    cluster-agent.yaml

    …
spec:
  …
appName: "my-cluster"
nsToMonitorRegex: ".*"
instrumentationMethod: "None"
containerAppCorrelationMethod: "none" 
targetAllocator:
    enabled: true         # false if autoScaling is enabled
    clusterAgentReplicas: 3
    autoScaling:
      enabled: false #false by default
      replicaProfile: Default 
      maxClusterAgentReplicas: 12
      scaleDown:
        stabilizationWindowSeconds: 86400 #In Seconds
    podConfig:
      imagePullPolicy: Always
      imagePullSecret: ecr-secret
      priorityClassName: high-priority
      serviceAccountName: appdynamics-target-allocator
      nodeSelector:{}
      tolerations:[]
      resources:
        limits:
          cpu: 500m
          memory: 500Mi
        requests:
          cpu: 200m
          memory: 200Mi
      labels:{}
      securityContext:{}
  9. Install the Cluster Agent:
    kubectl create -f cluster-agent.yaml
See Validate the Cluster Agent Installation to validate and troubleshoot the installation. Once the install is validated, see Container Installation Options for options to install App Server Agents in applications in the cluster.

Configuration to Use Custom Kubernetes API Server

To use Kubernetes API Server, you require to add the following to the cluster-agent-operator.yaml (for Kubernetes >=1.16) or cluster-agent-operator-openshift.yaml (OpenShift with Kubernetes >=1.16) spec :

  1. Update configMap in the cluster-agent-operator.yaml file with the Kubernetes API server details. These details are required:
    • server: specify the URL of the Kubernetes API server.
    • clusters.name: name of the cluster.
    • users.name: name of the user who can access the Kubernetes API server.
    • contexts.context: update the cluster name and user name to map the user details with the specific cluster.The following sample yaml file includes the required parameters:
      apiVersion: v1
kind: ConfigMap
metadata:
  name: custom-kubeconfig
  namespace: appdynamics
data:
  kubeconfig: |
    apiVersion: v1
    clusters:
    - cluster:
        server: "https://kubernetes.default.svc"
        certificate-authority: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt
      name: psr-cluster
    contexts:
    - context:
        cluster: psr-cluster
        user: psr-user
      name: psr-context
    current-context: psr-context
    kind: Config
    preferences: {}
    users:
    - name: psr-user
      user:
        tokenFile: /var/run/secrets/kubernetes.io/serviceaccount/token
      Here, https://kubernetes.default.svc is the URL of the API server. Under clusters, psr-cluster is the name of the cluster. Under users, psr-user is the user who can use the Kubernetes API server.
  2. Set the USE_CUSTOM_KUBECONFIG env variable to true in the cluster-agent-operator.yaml (for Kubernetes >=1.16) or cluster-agent-operator-openshift.yaml (OpenShift with Kubernetes >=1.16) file.
    
- name: USE_CUSTOM_KUBECONFIG
  value: "true