FAQ: Forever Incrementals - Introduction and Best Practices

Summary:

Forever Incremental removed the limit on the number of incremental backups allowed on a Windows File System backup. This new feature allows you to remove your full backup cycle and your backups will continue to run incremental backups.

This approach of backing up will remain recoverable if the full and its incremental chain remains in a healthy state. A new full backup is only required to reset the chain of incremental backups when breaks in the chain occur. This is most notable when the chain skips a number in the increment.

Special note for restores: The higher number of increments you must restore from, the longer the restore process will take. Keep this in mind when planning your backup strategy.

Solution:

Who should use this feature?

Forever Incremental is not well suited for everyone. The incremental will still take up storage space on your disk and in the cloud, so if you have large incremental backups, you could run out of storage. If your incrementals are small and you’re not changing much data on a day-to-day basis, then this can work for you.

Another situation where the Forever Incremental backups can be helpful is when you are running many incremental backups though out the day. However, when deploying this approach to backups It would be wise to do test restores of the highest increments, to ensure the restore times are tolerable for your needs.

Customers with low bandwidth may also find this feature helpful as they can greatly reduce the number of full backups that need to be taken. This will allow for reduced bandwidth usage over the long term.

Events that will require you to run a new full backup

There are events that will require that you do a new full backup to resolve them.

If you find that you are running out of storage, you will need to run a new full backup. Data retention is still set to your cycles and a full backup will start a new cycle for you. This does mean you will want to set your retention and make sure that you are scheduling/manually running your full backups when there is room for it to run.

The other event that will require a new full backup to be run is when you have a break in the chain of incremental backups. If you have a couple of errors, where it misses the incremental backup but does not have a skip in the number of the incremental such as having an incremental 5 then failing for a few days on incremental 6 but then finally getting a successful incremental 6, this is not a broken chain and does not require a new full backup. If you have an error and you go from incremental 5 to incremental 7 with incremental 6 having failed or missing, this is a broken chain and will require a new full backup.

Recommendations for using Forever Incrementals

All recommendations for retention and for the operation of your backup still apply for the safety of your backup:

• Retention of two: This will allow for reaction time if issues arise and the backup grabs bad data, for example backing up encrypted data. Even if you do not require a retention of two it is best to keep this and manually delete cycles after confirmation the new cycle is in good standing. See article Retention Policy Best Practices https://cbnt.it/1VQ1DiQ
• Email notifications: We recommend that you get your email notifications set up so that if there is an issue with your backup you will be notified about that right away. See article Configuring Email Notifications https://cbnt.it/2rkBr77
• Testing restores: As with any successful backup strategy, we highly recommend doing test restores of the data at a scheduled interval matching your needs.

Related Articles: