1. Download the images per the instructions in the Download location section, and make a note of the file names.
2. Log in to your Kubernetes cluster and set the context to the project/namespace for your existing deployment.
   $ oc login https://<cluster-ip>:<port> -u <cluster-admin> -p <password>
   $ oc project <existing deployment namespace>
3. When you have all of the files for the images that you want to install, run the following commands to get your access token and to make sure you can use Kubectl.
   $ oc whoami -t
   $ kubectl cluster-info
4. Check that you can run a  Podman command.
   $ podman ps -a
5. Get the registry route.

   $ oc registry info --public
   If the command has no output or the output is an internal service URL, it means that the route is not enabled. To enable the registry route on the cluster run the following command.
   $ oc patch configs.imageregistry.operator.openshift.io/cluster --type merge -p '{"spec":{"defaultRoute":true}}'
   For more information about exposing routes on OCP 4.6, see Exposing the registry. Use the OCP version menu to find your specific version in the OpenShift documentation.
6. Log in to the image registry by using the registry route that returns from the "oc registry info --public" command.
   $ podman login $(oc registry info --public) -u $(oc whoami) -p $(oc whoami -t) --tls-verify=false
   Note: If you are logged in to the cluster as "kubeadmin", the "oc whoami" command returns "kube:admin", which does not work. You must use "kubeadmin" as the login user.
   $ podman login $(oc registry info --public) -u kubeadmin -p $(oc whoami -t) --tls-verify=false
7. Change the permissions of the scripts/loadimages.sh script so that you can run it.
   $ cd scripts
   $ chmod +x loadimages.sh
8. Run the loadimages.sh script to load the images into your image registry. The following example shows the input values in the command line.
   $ ./loadimages.sh -p <ARCHIVE> -r $(oc registry info --public)/<project-name>
   Where:
   -p  The archive files location or archive file name
   -r  Target image registry and namespace
   Note: The <project-name> variable is the name of your existing deployment. Take a note of the image registry route so that you can enter it in the upgrade script. If you want to load the images into another project that can be referenced across namespaces, then you must allow pods to reference images from that project. For example, to allow any service account in a project that is named cp4a-project to reference images in another project named cp4a-images-project, use the oc policy add-role-to-group parameter.

   $ oc policy add-role-to-group \
      system:image-puller system:serviceaccounts:cp4a-project \
      --namespace=cp4a-images-project
   For more information, see Allowing pods to reference images across projects.

9. Check that the images are pushed correctly to the registry.
   $ oc get is
10. If you want to use an external registry, create a registry secret:
    $ oc create secret docker-registry admin.registrykey --docker-server=<registry_url> --docker-username=<your_account> --docker-password=<your_password> --docker-email=<your_email>
    Take a note of the secret and the server values so that you can set them to the "pullSecrets" and "repository" parameters when you update the operator for your containers.

1. Log in to your Kubernetes cluster and set the context to the project for your existing deployment.
   $ oc login https://<CLUSTERIP>:<port> -u <ADMINISTRATOR>
2. If the persistent volume (PV) for the operator is defined by a hostPath, provide group write permission according to the PV hostPath.path definition (/root/operator).
   $ chmod -R g=u /root/operator
   $ chmod g+rw /root/operator
   Note: If you are using dynamic provisioning, this step is not needed as the PV is created automatically as per the Storage Class definition.
3. Remove the .OPERATOR_TYPE file in case it exists from a previous deployment.
   $ rm -f /<hostPath>/.OPERATOR_TYPE
   Where hostPath is the value in your PV (root/operator).
4. Go to the downloaded cert-kubernetes folder for the IF002 interim fix.
5. Upgrade the operator in your project by running the following command.

   $ ./scripts/upgradeOperator.sh -i <registry_url>/icp4a-operator:20.0.3-IF002 -p '<my_secret_name>' -a accept

   Where registry_url is the value for your internal registry or cp.icr.io/cp/cp4a for the IBM Cloud Entitled Registry. The my_secret_name is the secret that is created to access the registry, and accept means that you accept the license.

   Note: If you plan to use a non-admin user to install the operator, you must add the user to the "ibm-cp4a-operator" role.

   $ oc adm policy add-role-to-user ibm-cp4a-operator <user_name>
6. Monitor the pod until it shows a STATUS of Running:
   $ oc get pods -w
   Note: When started, you can monitor the operator logs with the following command:
   $ oc logs -f deployment/ibm-cp4a-operator -c operator

Performing the necessary tasks after installation

Uninstalling

List of Fixes

Back to top

Back to top

Back to top

Back to top

Back to top

Back to top

Back to top

Back to top

Back to top

Back to top

Back to top

Back to top

Known Limitations