Uninstalling DataStage Enterprise

A project administrator can uninstall DataStage Enterprise from IBM® Cloud Pak for Data.

Before you begin

Required role: To complete this task, you must be an administrator of the project (namespace) where DataStage Enterprise is installed.

Before you uninstall DataStage Enterprise, ensure that the machine from which you will run the commands meets the following requirements:

  • Can connect to the cluster.
  • Has the oc command-line interface.
  • Has the Cloud Pak for Data command-line interface.

Removing DataStage Enterprise will remove all of the data that is associated with this instance of DataStage Enterprise.

Tip: For a list of all available options, enter the following command:
./cpd-cli uninstall --help

Procedure

From your installation node:

  1. Change to the directory where you placed the Cloud Pak for Data command-line interface.
  2. Log in to your Red Hat® OpenShift® cluster as a project administrator:
    oc login OpenShift_URL:port
  3. Run the following command to preview the changes that will be made to the cluster when you remove DataStage Enterprise:
    ./cpd-cli uninstall \
--assembly ds-ent \ 
--namespace Project \
--include-dependent-assemblies \ 
--dry-run

    Replace the following values:

    Variable Replace with
    Project The Red Hat OpenShift project where DataStage Enterprise is deployed.
    Important: Review the output of the command to understand what software will be removed.

    The --include-dependent-assemblies flag automatically removes any dependencies that are not required by other services.

    For example, if DataStage Enterprise is the only service installed on Cloud Pak for Data, this command will uninstall DataStage Enterprise and the Cloud Pak for Data control plane.

    In this scenario, if you don't want to uninstall the Cloud Pak for Data control plane, you can remove the --include-dependent-assemblies flag from the command.

  4. Rerun the previous command without the --dry-run flag to remove DataStage Enterprise.

What to do next

If you want to reuse the cluster for deployment, complete the following cleanup steps:
  • Delete all PersistentVolumeClaim (PVC) and PersistentVolume (PV) resources:
    oc get pvc -n ds-ent-namespace
oc delete pvc ds-ent-pvname -n ds-ent-namespace
oc get pv -n ds-ent-namespace
oc delete pv ds-ent-pvname -n ds-ent-namespace
  • Delete all job resources:
    oc get jobs -n ds-ent-namespace
oc delete job ds-ent-jobname -n ds-ent-namespace
  • Delete all cronjob resources:
    oc get cronjob -n ds-ent-namespace
oc delete cronjob ds-ent-cronjobname -n ds-ent-namespace
  • If you did not deploy Watson™ Machine Learning or Watson Studio, delete the config map resources:
    oc get configmap -n ds-ent-namespace
oc delete configmap ds-roles-config -n ds-ent-namespace
oc delete configmap ds-ent-config -n ds-ent-namespace
  • Delete the completed DataStage Enterprise pods:
    oc delete pod -n <namespace> `oc get pod -n <namespace> | grep ds- | grep Completed | awk '{print $1}'`