Before you upgrade

Copy link

• Ensure the application is in a stable state and that you have documented any critical configuration values currently in use.
• A current backup of the PostgreSQL database for Red Hat Edge Manager (the flightctl database) that meets your organization’s recovery point objective, or a maintenance window to create that backup before you upgrade.
• Back up Red Hat Edge Manager on the cluster before you change the Helm release. Use Backing up the full server on Red Hat OpenShift Container Platform (flightctl-backup), or follow Backing up the PostgreSQL database for Red Hat Edge Manager on Red Hat OpenShift Container Platform if your runbook covers the database only.

Upgrading with the Helm CLI (recommended)

Copy link

Use this method when you are changing the chart version and want every value from the current release carried forward and merged with the new chart’s defaults. Replace <release_name>, <namespace>, and <target_version> with your Helm release name, OpenShift project, and the chart version to install.

• The Helm CLI (version 3.x) is installed.
• You are authenticated to the cluster (oc login or a kubeconfig with access to the namespace).
  1. Add the OpenShift Helm chart repository and refresh the local chart index:

     $ helm repo add openshift-helm-charts https://charts.openshift.io/
     $ helm repo update

  2. Run the upgrade using the openshift-helm-charts/redhat-rhem chart reference and --reset-then-reuse-values:

     $ helm upgrade <release_name> openshift-helm-charts/redhat-rhem \
         --version <target_version> \
         --namespace <namespace> \
         --reset-then-reuse-values

     The --reset-then-reuse-values flag resets values to the new chart version’s defaults, then reapplies values from the installed release so your custom settings persist across the upgrade.

  3. Verify that custom values are still present:

     $ helm get values <release_name> -n <namespace>

     Example output:

     USER-SUPPLIED VALUES:
     global:
       auth:
         caCert: |
           -----BEGIN CERTIFICATE-----
           ...
           -----END CERTIFICATE-----
       enableMulticlusterExtensions: "false"

     Your output lists the values stored on the release. Confirm that fields you customized—such as global.auth.caCert—are still present.

Upgrading from the OpenShift web console

Copy link

Use this method if you cannot use the Helm CLI. When you change the chart version, you must merge any custom values yourself; the editor does not pre-fill them from the running release.

• You are logged into the Red Hat OpenShift Container Platform web console with permission to modify resources in the namespace where Red Hat Edge Manager is installed.
• The correct Project is selected.
• You have a record of every Helm value that differs from the chart defaults. You can obtain these values by running helm get values <release_name> -n <namespace> on a workstation with cluster access, or by viewing the installed release’s YAML in the console before you start the upgrade.
  1. In the OpenShift console, go to Helm Releases, find your Red Hat Edge Manager release, open the kebab menu (three vertical dots) on the row, and select Upgrade.

     Alternatively, open the upgrade flow from Ecosystem Software Catalog for the installed Red Hat Edge Manager release, if that path is available in your console version.

  2. If you are moving to a new chart version, select it from the Chart Version menu.
  3. In the YAML editor (Values), merge the custom values you recorded in the prerequisites with the displayed YAML. Do not rely on the editor alone to preserve settings such as global.auth.caCert when the chart version changes.
  4. Click Upgrade to deploy.
  5. Verify the release. Because you upgraded through the web console in this procedure, inspect Helm Releases in the console and confirm the deployed release reflects your custom settings.

Once the upgrade runs, you can watch the Status column on Helm Releases change to Pending and then Deployed when the update succeeds.

If Operators in your organization previously wrote Catalog or CatalogItem resources directly, review Restoring catalog write access for Operator accounts after you upgrade.