Before you upgrade your Cisco Nexus Dashboard Orchestrator cluster:

• Ensure that you are running Nexus Dashboard Orchestrator release 3.3(1) or later.

  Note

  If you are running a release prior to 3.3(1), you must skip this chapter and follow the instructions described in Upgrading Manually Using Configuration Restore instead.

• Note that downgrading from this release is not supported.

  If you ever want to downgrade, you can deploy a brand-new cluster using the earlier version and then restore configuration from the earlier release. Note that you cannot restore a backup created on a newer version in an older version, in other words restoring a backup from release 4.2(1) in release 3.7(1) is not supported.

• Ensure that your current Nexus Dashboard cluster is healthy.

  You can check the Nexus Dashboard cluster health in one of two ways:

  • By logging into your Nexus Dashboard GUI and verifying system status in the System Overview page.

  • By logging into any one of the nodes directly as rescue-user and running the following command:

    # acs health
    All components are healthy

• Ensure that your current Cisco Nexus Dashboard Orchestrator is healthy.

  Depending on your existing Orchestrator version, you can check the service status in one of two ways:

  • For recent NDO releases, check the Nexus Dashboard's Overview page that shows the platform and services' health.

  • For older Orchestrator releases, you can check the status of you Nexus Dashboard Orchestrator service by navigating to Settings > System Status:

    

    Then ensure that the status of all nodes and services is healthy:

    

• Ensure that there are no configuration drifts before you back up your existing configuration.

  This applies to all template types available in your existing release, such as application, tenant policies, fabric policies, and fabric resource policies templates.

  If your existing Nexus Dashboard Orchestrator is release 3.7(1) or later, you can use the drift reconciliation workflow for application templates, as described in the "Configuration Drifts" section of the Nexus Dashboard Orchestrator Configuration Guide.

• Back up and download your existing Orchestrator configurations.

  Configuration backups are described in the "Backup and Restore" chapter of the Nexus Dashboard Orchestrator Configuration Guide for your release.

• Back up and download your existing fabrics' configurations.

  We recommend running configuration drift reconciliation after you upgrade your Nexus Dashboard Orchestrator, which may require you to redeploy configurations to your fabrics. As such, we recommend creating configuration backups of all fabrics managed by your Nexus Dashboard Orchestrator.

  For more information on creating Cisco APIC configuration backups, see the "Management" chapter of the Cisco APIC Basic Configuration Guide for your release.

  For more information on creating Cisco Cloud Network Controller configuration backups, see the "Configuring Cisco Cloud Network Controller Components" chapter of the Cisco Cloud Network Controller User Guide for your release.

  For more information on creating Cisco Nexus Dashboard Fabric Controller configuration backups, see the "Backup and Restore" chapter of the Cisco NDFC Fabric Controller Configuration Guide for your release.

• Note that if you have template versioning enabled (supported since release 3.4(1)), only the latest versions of the templates are preserved during the upgrade.

  All other existing versions of templates, including older versions that are tagged Golden, will not be transferred.

• Ensure that all templates are in a supported state before creating the configuration backup of the existing cluster:

  • Templates that are undeployed or were never deployed after they were created require no additional attention and will be migrated during the upgrade.

  • All deployed templates must have no pending configuration changes.

    If you have one or more templates that have been modified since they were last deployed, you must either deploy the latest version of the template or undo the changes to the template since it was deployed by reverting to the last-deployed version and re-deploying it.

• When upgrading the Orchestrator service, you can do so in one of two ways:

  • Using the Nexus Dashboard's App Store, as described in Upgrading Orchestrator Service Using Cisco App Store.

    In this case, the Cisco DC App Center must be reachable from the Nexus Dashboard via the Management Network directly or using a proxy configuration. Nexus Dashboard proxy configuration is described in the Nexus Dashboard User Guide.

    Note

    The App Store allows you to upgrade to the latest available version of the service only. If you want to upgrade to a different release, you must use the manual upgrade process as described below.

  • By manually uploading the new app image, as described in Upgrading Orchestrator Service Manually.

    You can use this approach if you are unable to establish the connection to the DC App Center or if you want to upgrade to a version of the service that is not the latest available release.

• SR-MPLS and SDA integration configurations are not transferred during the upgrade.

  If you have either of these integrations in your deployment, it will not affect the migration, but you will receive a notification and will need to reconfigure them after you complete the upgrade.

• If you plan to add and manage new Cloud Network Controller sites after you upgrade your Nexus Dashboard Orchestrator to this release, ensure that they are running Cloud Network Controller release 5.2(1) or later.

  On-boarding and managing Cloud Network Controller sites running earlier releases is not supported.