There are three stages in the upgrade process of vRealize Automation within vRealize Suite Lifecycle Manager.

Stages of Upgrade Description

Upgrade Preparation

The preparation phase verifies that the system is healthy and shuts down services to make sure that all data is persisted.
Snapshot of all of the VMs for an automatic recovery Snapshots are taken for faster recovery of failures. vRealize Suite Lifecycle Manager then shuts down the VMs, takes a snapshot, turns the power on, and continues to the next phase.
Upgrade Execution Execute the upgrade.
At certain unexpected or failure scenarios, the vRealize Suite Lifecycle Manager upgrade workflow gives user decision points to either complete the upgrade successfully or revert to the stage before upgrade.
  • The upgrade process starts with a status check task that verifies the current state of the VA. If the system already has an upgrade request due to a previous upgrade attempts, then vRealize Suite Lifecycle Manager provides you with an option to clean the older states and start a new upgrade. You can see the status task failing with a retry parameter similar to 'cancelAndStartAfresh'. Setting this retry parameter to true cleans up older states and automatically retriggers the upgrade again.
  • If unexpected failures occur during the prepare phase, you are provided with an option to either cancel the whole upgrade process and start new if a failure can never be corrected or if the unexpected failure is fixed manually, you can proceed to the next phase in the upgrade workflow. The status check done after the prepare failure provide you with two retry parameters that help you decide which option to choose. If you set 'cancelAndStartAfresh' to 'true', the upgrade process is cancelled and returns the machine to the state before the upgrade. If you set ‘proceedNext’ to 'true', the vRealize Suite Lifecycle Manager upgrade workflow proceeds to the next state with an assumption that you fixed the prepare issue outside.
  • The upgrade workflow consists of VM level operations like taking, reverting, deleting a snapshot and VM Shutdown, power ON, and so on. If there was failure, these operations includes a Skip option which can be used if the RETRY option in the vRealize Suite Lifecycle Manager does not help and when you manually perform the same operation directly on the vCenter Server.
  • The final phase of the upgrade can be successful or success with warnings or a fatal state.
    • Success with warnings indicates that the upgrade has completed successfully, but a minor error is detected. You can check the errors and rectify them. You can set the 'succeedUpgradeRequest' retry parameter to 'true' which will succeed and complete the vRealize Suite Lifecycle Manager upgrade workflow.
    • If there is fatal error, you can decide if you want to revert the snapshot and retry the upgrade or cancel the whole upgrade process. This reverts and deletes the snapshot, and cancels the current upgrade request and moves the system to a state before the upgrade started.
      • For a fatal error, you can see the status task after the upgrade failure with retry parameters similar to 'revertSnapshotNRetryUpgrade' and 'cancelUpgradeNRevertBack'. If you set 'revertSnapshotNRetryUpgrade' to 'true', then it can revert the snapshot and retry the upgrade again.
      • If you set 'cancelUpgradeNRevertBack' to 'true', you can cancel the upgrade process, which can revert and delete the snapshot and cancel the current upgrade request and move the system to a state before the upgrade started.
Note:
  • The vRealize Suite Lifecycle Manager upgrade workflow does not support removing the snapshots if there was a successful upgrade. You hold the snapshots or remove it manually from the vCenter Server if necessary.
  • If you Cancel the upgrade process after a failure post prepare or upgrade phase, the vRealize Suite Lifecycle Manager workflow post cleaning up the upgrade request in vRealize Automation ends up in an aborted state. This indicates that the upgrade workflow from vRealize Suite Lifecycle Manager was aborted. In such situations, retrigger an upgrade from the product actions under Manager Environment page.