Upgrading Netcool Operations Insight on Red Hat OpenShift using the Operator Lifecycle Manager (OLM) user interface

Edit online

Use these instructions to upgrade an existing Netcool® Operations Insight® deployment from version 1.6.4 to 1.6.5, using the Red Hat® OpenShift® Operator Lifecycle Manager (OLM) user interface (UI).

Before you begin

Note: Upgrading cloud-based deployments from version 1.6.3.2 is not supported. You must first upgrade IBM® Netcool Operations Insight on Red Hat OpenShift to version 1.6.4 and then upgrade to version 1.6.5. For more information, see Upgrading and rolling back on Red Hat OpenShift in the version 1.6.4 documentation.
  • Ensure that you have completed all the steps in Preparing your cluster. Most of these steps should already have been completed as part of your previous Netcool Operations Insight deployment.
  • Ensure that you have an adequately sized cluster. For more information, see Sizing for a Netcool Operations Insight on Red Hat OpenShift deployment.
  • Your deployment must have persistent storage configured. Only version 1.6.4 deployments with persistence enabled are supported for upgrade to version 1.6.5.
  • Before you upgrade to version 1.6.5, if present, remove the noi-root-ca secret by running the following command:
    oc delete secret noi-root-ca
  • Before you upgrade to version 1.6.5, if present, reverse any image overrides from the hot fix of the previous release.
    1. Edit the custom resource (CR):
      oc edit noi <release-name>
      Where <release-name> is the release name, for example, evtmanager.
    2. Manually remove the tag, name, and digest entries of image overrides from the helmValuesNOI section of the YAML file.

If you want to verify the origin of the catalog, then use the OLM UI and CASE upgrade method instead. For more information, see Upgrading a cloud deployment offline using the OLM UI and CASE (airgap).

All the required images for version 1.6.5 are either in the freely accessible operator repository (icr.io/cpopen), or in the IBM Entitled Registry (cp.icr.io) for which you will require an entitlement key.

Procedure

Upgrade the Catalog source

  1. From the Red Hat OpenShift OLM UI, navigate to Administration > Cluster Settings, and then select the OperatorHub configuration resource under the Global Configurations tab.
  2. Under the Sources tab, click the existing Netcool Operations Insight catalog source.
  3. Edit the catalog source YAML and replace spec.image with the Netcool Operations Insight catalog source name and image for version 1.6.5, icr.io/cpopen/ibm-operator-catalog:latest. Select the Save button.

Upgrade the Netcool Operations Insight operator

  1. To upgrade the operator for OLM UI, navigate to Operators > Installed Operators. Go to the Subscription tab and select Update channel to v1.9 for 1.6.5.1. If you are upgrading to 1.6.5, select Update channel to v1.8.
    Note: It takes a few minutes for IBM Cloud Pak for Watson™ AIOps Event Manager to install. When installed, ensure that the status of IBM Cloud Pak for Watson AIOps Event Manager is Succeeded before you proceed to the next steps.

Upgrade the Netcool Operations Insight instance

  1. From the Red Hat OpenShift OLM UI, upgrade the Netcool Operations Insight instance. Go to Operators > Installed Operators and select your Project. Then select IBM Cloud Pak for Watson AIOps Event Manager.
  2. Go to the All instances tab and select your instance. Edit the Netcool Operations Insight instance YAML.
    version 1.6.5.1 iconUpgrading to 1.6.5.1: Update spec.version: 1.6.5 to spec.version: 1.6.5.1.
    version 1.6.5.0 iconUpgrading to 1.6.5: Update spec.version: 1.6.4 to spec.version: 1.6.5.
  3. Select the Save button.
  4. The logstash pod is replaced with the common-datarouting pod in version 1.6.5. Delete the logstash and spark deployments by running the following commands: 
    oc delete deployment <release-name>-logstash
oc delete deployment <release-name>-spark-master
oc delete deployment <release-name>-spark-slave
  5. After you restart the noi-operator, confirm that the logstash and spark deployments are not recreated.

What to do next

If the <release name>-ibm-hdm-analytics-dev-dedup-aggregationservice pod does not start up, restart the redis pods. For more information, see Service does not start up after upgrade.