Service protection

The IBM Storage Fusion Backup & Restore service protection involves the backup of control plane to a S3 bucket. In the event of cluster failure, you can use this feature to restore the Backup & Restore service to another cluster. Configure service protection and run the initial service backup.

Service protection is configured on one cluster, with both application and service backups run. Then the same bucket on cloud storage is used to configure service protection on another cluster with the intent to restore service and application backups from the first cluster. The following may result in backups appearing in the UI that no longer exist on cloud storage, and any restore attempts using those backups fail until this is fixed.

  1. If the first cluster remains as-is, the retention period on the original backups may expire and backups may be removed them from cloud storage. As the second cluster is unaware of the removal, when it attempts to remove the backups that had been restored, the backup information on the cluster does not get removed, because the backup on cloud storage no longer exists.
  2. If Backup & Restore is uninstalled from the first cluster, the -s option should be used to prevent DeleteBackupRequest CRs from being created. If the option is not set, the backups on cloud storage will be removed. Again, the second cluster will be unaware that they no longer exist on cloud storage.

These first two may be considered configuration errors, since service protection is not meant as a migration tool. The first cluster's deployment should no longer exist when the second cluster is configured.

  1. If a service backup - that is not the latest - is restored, it may contain application backups that have since expired. The likelihood varies depending on how often application and service backups are taken, as well as their respective retention settings.