Installing the IBM App Connect Operator from the Red Hat OpenShift web console

Use these instructions to install the IBM® App Connect Operator by using the Red Hat® OpenShift web console.

To install and subscribe to the IBM App Connect Operator, complete the following steps:

1. Log in to the OpenShift web console for your cluster.
2. If using Red Hat OpenShift 4.19 or earlier, ensure that you are in the Administrator perspective .
3. If you want to install the IBM App Connect Operator into an existing namespace or cluster wide, go to step 4. Otherwise, create a namespace to use for the IBM App Connect Operator as follows:
   1. From the navigation, click Home > Projects to open the Projects page.
   2. Click Create Project.
   3. Enter a namespace (or project) name; for example, ibm-appconn. Optionally add a display name and description.
   4. Click Create.
   Restriction: Do not install into any of these system namespaces, which might impose restrictions on assigning role-based access control (RBAC) permissions or security context constraints:

   • The default namespace
   • Any namespace prefixed with kube-
   • The openshift namespace or any namespace prefixed with openshift-

     The exception is the openshift-operators namespace, which is used for a cluster-wide installation.

4. Install the IBM App Connect Operator:
   1. From the navigation, click Ecosystem > Software Catalog if using Red Hat OpenShift 4.20 or later, or click Operators > OperatorHub if using Red Hat OpenShift 4.19 or earlier.
   2. From the Software Catalog or OperatorHub page, search for app connect, and then click the IBM App Connect tile.
      
   3. Click Install.
   4. From the Install Operator page, complete the fields as follows:
      • Update channel: The supported update channels are shown, with v12.21 selected by default. This indicates that an Operator subscription will automatically be created to keep the Operator up to date when new versions are delivered to the channel.
      • Installation mode: Choose whether to install the Operator into all namespaces in the cluster or into a specific namespace.
      • Installed Namespace: If you chose the option for a specific namespace, select a namespace from this list.
      • Update approval: Click Automatic (the default) to indicate that the installation should proceed with no additional approval. This option will also cause the running instance of your Operator to be automatically upgraded whenever new versions are delivered to the channel.

        Click Manual if you want to review a generated Install Plan for the Operator and then manually approve the installation. This option will also require you to review the Install Plan for each new Operator version that is delivered to the channel, and then manually approve an upgrade. Do not choose this option if you are installing the Operator from a product-specific App Connect catalog source. The use of a product-specific catalog source enables you to control which upgrades to apply because upgrades happen only when the image digest for the catalog source is changed.

        Tip: If required, you can change the approval strategy later from the Subscription tab for the installed IBM App Connect Operator. You can access this tab by clicking Ecosystem > Installed Operators or Operators > Installed Operators in the navigation pane, and then clicking IBM App Connect.

        
   5. Click Install to install the Operator.

      If you chose a Manual approval strategy, you are required to review and approve the Install Plan of the subscription. No additional action is needed if you selected an Automatic approval strategy.

      Wait for the installation to complete and then click View installed Operators in Namespace. The IBM App Connect Operator is displayed on the Installed Operators page with a deployment status of Succeeded / Up to date.

      
   6. Click IBM App Connect to view details about the APIs that are provided.
      
   Deploying the Operator with additional configuration options:

   On Red Hat OpenShift, Operator Lifecycle Manager (OLM) runs by default to aid with the installation, update, and lifecycle management of Operators and services that run in the cluster. When OLM deploys the IBM App Connect Operator, a Subscription object is automatically created for that Operator. This Subscription object identifies the name and namespace of the Operator, the catalog source for the Operator, the channel that OLM tracks for new Operator versions, and whether Operator upgrades should be automatic or manual.

   You can additionally configure how OLM should deploy the Operator by using the config field in the spec section of the Subscription's custom resource (CR). You can apply this configuration only by using the Red Hat OpenShift CLI to update the CR.

   IBM App Connect Operator 12.7.0 or later enables you to configure scheduling options for the Operator pod that is created by OLM. You can apply nodeSelector settings to constrain your Operator pod to nodes that are assigned specific labels, and apply tolerations (in conjunction with taints) to ensure that the Operator pod is not scheduled onto inappropriate nodes. To apply this configuration, you can add the following parameters to the Operator's Subscription CR:

   • spec.config.nodeSelector: Use this parameter to specify a set of key/value pairs that must be matched against the node labels to decide whether the Operator pod can be scheduled on that node. Only nodes whose labels match all of these key/value pairs will be selected for scheduling the Operator pod.
   • spec.config.tolerations.effect, spec.config.tolerations.key, spec.config.tolerations.operator, spec.config.tolerations.tolerationSeconds, and spec.config.tolerations.value: Use these parameters to apply an array of one or more tolerations to prevent the Operator pod from being scheduled on inappropriate nodes.

     First, apply one or more taints to a node (by running oc taint with a key, value, and taint effect) to indicate that the node should repel any pods that do not tolerate the taints. Then, use the spec.config.tolerations.* parameters to apply toleration settings (effect, key, operator, toleration period, and value) to the pod to allow it to be scheduled on the node if the pod's toleration matches the node's taint.

   For more information, see nodeSelector, Assign Pods to Nodes, and Taints and Tolerations in the Kubernetes documentation.

   You can configure scheduling options for your Operator pod by updating the Subscription CR for the IBM App Connect Operator as follows:

   1. From the command line, log in to your cluster by using the oc login command.
   2. Run the oc edit command to partially update the subscription, where namespaceName is the namespace where the Operator is installed.

      oc edit subscription ibm-appconnect -n namespaceName

      The YAML manifest (or CR) for the Subscription object automatically opens in the default text editor for your operating system.

   3. Update the spec section of the file to add or update the config field with your nodeSelector or tolerations settings. For example, if you applied workloadType=ace:NoSchedule as a taint and workloadType=ace as a label to a worker node, you can add a matching nodeSelector and toleration to schedule the Operator pod onto the tainted worker node:

      spec:
        channel: v12.21
      ...
        config:
          nodeSelector:
            workloadType: ace
          tolerations:
            - key: workloadType
              operator: Equal
              value: ace
              effect: NoSchedule
              tolerationSeconds: 100
      ...

   4. Save the YAML definition and close the text editor to apply the changes.

   You can also configure other options such as volumes, volume mounts, and resources in a similar way. For information about all the spec.config settings that you can configure for an Operator's Subscription object, see Configuring Operators deployed by OLM in the Operator Lifecycle Manager documentation.