A data retention policy is an agreement on retaining data for operational or regulatory compliance needs.
A data retention policy appears as a part of an overall data management and plays a significant role since the definition of the terms of keeping a company's data is crucial. Data retention for longer periods than needed produces unnecessary storage usage and increases storage costs.
If user data is modified or deleted during the retention period, a copy of the original content can be always found in the backup storage.
Objects in backup storage whose retention period has expired, are deleted automatically, according to retention policy settings, so this keeps the backup storage size under control. The duration of the data retention policy can be ranged from days to years.
This chapter covers the following topics:
Retention Policy In the Current Backup Format
This paragraph describes settings available for the retention policy for the current backup format.
On this tab, you can specify the global retention policy settings that will apply to all backup plans by default, unless they have been provided with individual retention policy settings.
The following retention policy settings are available on this tab:
- You can make a backup service delete file versions that were modified or uploaded to your backup storage more than a certain number of days ago. When enabling this option, you can also make the backup service always keep the last version of each file regardless of when they were backed up.
- You can specify the maximum number of versions that a backup service should keep in the storage for each file.
- You can specify a time period for which locally deleted files should be maintained in the backup storage before being purged. After these files become removed from the backup storage, you will no longer be able to restore them.
In addition, you can specify separate retention settings applied for the history that MSP360 (CloudBerry) Backup maintains on the History tab to keep track of any changes made to the local and target storages, as well as the local repository. The following retention policies are available for the backup history:
- You can make the history keep records only for a certain period of time. Records will be deleted once they exceed the specified lifespan.
- You can limit the maximum number of records that history can maintain.
With Amazon S3 and Google Cloud, you can configure lifecycle rules that define when and how to delete your data or move it to another storage class
Retention Policy In the New Backup Format
The data retention approach in the new backup format is different as data is always purged by generations.
Generation in the new backup format terms is a sequence of full and incremental backups.
For example, if your backup plan schedule is set in such a way that an incremental backup is executed daily, and full backup is executed once a week, the generation will contain one full backup and six incremental backups for this backup plan.
Thus, the retention policy for the new backup format is always applied to the one backup plan.
In fact, the retention policy has only one setting: a period of backup keeping. The retention policy can be modified on the backup plan pane in the Backup Plans tab or via backup plan editing.
After the specified period, the backup data will be purged on backup storage.
Backup Data Purge On Backup Storage
Purge works as a cycle that goes around every generation. A successful restore point is searched for for the current generation. If no successful restore point is found, then the previous generation is checked. If a valid restore point in the generation being processed, then it is considered as the starting date of the retention period.
Purge can be seen in the console under the Backup Storage tab. If the retention period is specified, the corresponding legend is displayed to the right of the bunch.
If 30 days or less are left before the purge date, the number of days is displayed. If 10 or fewer days are left before the purge, the notification is red.
Additionally, the backup tree displays generations were set to delete and were not deleted for some reason. Purge interrupted notification is displayed for such nodes
An automatic purge cycle is started at the end of the backup plan execution as the backup data is uploaded. Algorithmically, purge consists of two stages:
- Generations are marked to purge in accordance with the retention policy of the plan
- Delete operations for all generations of the current plan are created.
Restore points created in previous versions are not included in automatic purge.
Regardless of the retention policy settings, the bunch must contain at least one current generation. The generation that has a valid restore point is considered as current.
The current generation is never a subject for purge