Dealing with Corrupted Recovery Points

Product version: 9.3

Last modified: 25 May 2020

Problem

One or more recovery points in my Backup Repository are corrupted.

  • Why does the repository corruption happen?

  • What can I do to prevent corruption in the future?

Background

A recovery point can become corrupted if one or more data blocks related to this recovery point is/are invalid or missing. This can happen due to the following:

  • Powering down or rebooting the Transporter during space reclaim task

  • Storage issues (bad blocks, physical storage corruption)

  • OS issues, for example, the OS may report that data was written while it was not physically written (delayed write, cache, and so on)

  • If there are corrupted recovery points in a Backup Repository and you continue using it, the number of corruptions may increase.

If backup verification finds that a recovery point is corrupted, a red icon will be displayed over the affected backup in the Backup Repository details. To view which recovery points are affected, expand the backup. Corrupted recovery points are marked with a yellow icon. If a recovery point is marked as corrupted, you still can try to recover files from it or run Performing Flash VM Boot Recovery. NAKIVO Backup & Replication does not prohibit recovery from such recovery points, although successful recovery cannot be guaranteed.

Solution

Since corrupted recovery points may affect new data or data from other backups/recovery points, we recommend you do the following:

  • If the last recovery point is corrupted, run the job for the backup with the corrupted recovery point. NAKIVO Backup & Replication automatically rebuilds the last recovery point using the data from the source VM.

  • If the corrupted point is not the last one:

    1. Delete the corrupted recovery point.

    2. Verify backups in the Backup Repository.

    3. If one or more corrupted recovery points appear, repeat the above process.

Important
If your only recovery point for a VM is corrupted, you can’t delete it.

After you remove the corrupted recovery points,backup verification will ensure other recovery points are good. We also recommend that you schedule backup verification so it runs periodically.

Note
Maintenance tasks, such as space reclaim, self-healing, or another backup verification session will clear the previous verification results.

In order to prevent corruptions in the future, we recommend the following:

  • Make sure that the Transporter assigned to the Backup Repository cannot be powered off or rebooted unexpectedly.

  • If possible, use a couple of smaller repositories instead of a single large Backup Repository. This way you can keep them more manageable: space reclaim and repository verification will happen faster.

  • To avoid disrupting NAKIVO Backup & Replication processes and to avoid data corruption, add the application to the whitelist/exclusions list of the antivirus software running on the machine on which the NAKIVO Backup Repository is set up.