Deploying the API Analytics subsystem on Linux® x86_64 (CLI)

Log in to your cluster with your OpenShift user credentials:

oc login

If you installed the operators in All namespaces on the cluster mode, you need to use a project other than openshift-operators in which to deploy the instances.

• If needed, create a new project in which to create the Operand Request object:

  oc new-project <project_name>

  For example:

  oc new-project integration

Create a file called operand-request.yaml and add the following content:

apiVersion: operator.ibm.com/v1alpha1
kind: OperandRequest
metadata:
  name: ibm-apiconnect-cert-manager
spec:
  requests:
  - operands:
    - name: ibm-cert-manager-operator
    registry: common-service
    registryNamespace: ibm-common-services

Create the resource:

oc apply -f operand-request.yaml

Change to the namespace where you want to install the subsystem:

oc project <namespace>

If you have installed the API Manager subsystem in another namespace, extract the API Manager ingress-ca certificates.

oc -n <api manager namespace> get secret ingress-ca -o yaml > ingress-ca.yaml

Edit the ingress-ca.yaml file and remove the following properties:

• metadata.creationTimestamp

• metadata.namespace

• metadata.resourceVersion

• metadata.uid

• metadata.selfLink Keep this file. You will need to apply it in the namespaces of your other subsystems so that they can communicate with the API Manager subsystem.

Apply the file to the API Analytics namespace:

oc apply -f ingress-ca.yaml

Create a file that is called api-analytics-certs.yaml and paste in the following contents:

apiVersion: cert-manager.io/v1
kind: Issuer
metadata:
  name: selfsigning-issuer
  labels: {
    app.kubernetes.io/instance: "api-manager",
    app.kubernetes.io/managed-by: "ibm-apiconnect",
    app.kubernetes.io/name: "selfsigning-issuer"
  }
spec:
  selfSigned: {}
---
apiVersion: cert-manager.io/v1
kind: Issuer
metadata:
  name: ingress-issuer
  labels: {
    app.kubernetes.io/instance: "api-manager",
    app.kubernetes.io/managed-by: "ibm-apiconnect",
    app.kubernetes.io/name: "ingress-issuer"
  }
spec:
  ca:
    secretName: ingress-ca

Apply the file to your namespace:

oc apply -f api-analytics-certs.yaml

Verify that the command installation succeeded:

oc get issuers

All issuers created successfully:

NAME                 READY
ingress-issuer       True
selfsigning-issuer   True

Create a AnalyticsCluster YAML file. For example, you can create a file that is called api-analytics.yaml with the following example configuration. All fields in the example are required. Update the values as applicable for your configuration:

apiVersion: analytics.apiconnect.ibm.com/v1beta1
kind: AnalyticsCluster
metadata:
  name: api-analytics
  labels: {
    app.kubernetes.io/instance: "api-analytics",
    app.kubernetes.io/managed-by: "ibm-apiconnect",
    app.kubernetes.io/name: "api-analytics"
  }
  annotations: {
    apiconnect-operator/cp4i: "false"
  }
spec:
  version: 10.0.6.0
  license:
    accept: false
    use: production
    license: L-KZXM-S7SNCU
  profile: n1xc2.m16
  microServiceSecurity: certManager
  certManagerIssuer:
    name: selfsigning-issuer
    kind: Issuer
  ingestion:
    endpoint:
      annotations:
        cert-manager.io/issuer: ingress-issuer
      hosts: 
      - name: ai.$STACK_HOST
        secretName: analytics-ai-endpoint
    clientSubjectDN: CN=analytics-ingestion-client,O=cert-manager
  storage:
    type: shared
    shared:
      volumeClaimTemplate:
        storageClassName: <storage-class>
        volumeSize: 50Gi

• Change the value of spec.license.accept to true if you accept the license agreement. For more information, see Licensing.

• Do not remove the annotation of apiconnect-operator/cp4i: "false". This annonation guarantees that API Analytics does not attempt to integration with the Platform UI. This integration is not supported.

• For spec.license.use enter production or nonproduction to match the type of license that you purchased.

• For spec.license.license enter the license ID for the API Connect program that you purchased. To get the available license IDs, see API Connect licenses in the API Connect documentation.

• For spec.profile enter the type of installation profile that you want. For more information, see API Connect deployment profiles for OpenShift and Cloud Pak for Integration.

• For spec.version enter the API Connect product version or channel to be installed.

• For <storage-class>, specify the RWO block storage class to use for persistence storage. For more information about selecting storage classes for Cloud Pak for Integration, see Storage considerations. To review API Connect storage support, see the "Supported storage types" section in Deployment requirements, in the API Connect documentation.

• Replace $STACK_HOST with the desired ingress subdomain for the API Connect stack. This variable is used when specifying endpoints. Domain names that are used for endpoints cannot contain the underscore "_" character. The host on OpenShift is typically prefixed by apps. Such as apps.subnet.example.com.

• If you want to enable additional analytics features, you can set them as described in Analytics subsystem CR configured settings.

Optional: Three replica deployments only: Enable dedicated storage. If you want to use dedicated storage, update the spec.storage section of the yaml file as follows:

storage:
  type: dedicated
  shared:
    volumeClaimTemplate:
      storageClassName: <storage-class>
      volumeSize: 50Gi
  master:
    volumeClaimTemplate:
      storageClassName: <storage-class>

For more information about dedicated storage, see Dedicated storage instead of shared storage.

Apply the YAML file to the cluster:

oc apply -f api-analytics.yaml

Check the status of API Analytics by running the following command in the project (namespace) where it was deployed:

oc get AnalyticsCluster

The installation is complete when the READY status changes to True, and the SUMMARY reports that all services are online:

NAME         READY   SUMMARY   VERSION    RECONCILED VERSION       AGE
api-analytics   True   6/1       <version>   <version-build>       7m17s