S3 AWS Backup - Understanding scheduling & incremental & continuous backup

Question

Hello Community !

Let's say I have a Continuos Backup for S3.  It is scheduled on a daily basis with a retention of 35 days.
In this way I would have two strategies. Every day I would be doing a scheduled incremental backup + the continuous backup with the changes made in the bucket until the next day.

In my mind, it should create a recovery point in my vault every day, and each of these recovery points would have a 24 hour restore point. However, even though the backup jobs are completed successfully. There are days when a new recovery point is not created, but I have a recovery point with a recovery window of 3 days for example.
So I don't see consistency in this process.

I have another continuous backup for RDS. It is scheduled on a daily basis with a 35 day retention, but in this case I have a single recovery point, with a 35 day restore point window that keeps moving.

Any explanation for this behavior of the S3 backup ? How does the scheduling work ? If I don't specify any schedule, a default one is set.

According to "[https://repost.aws/questions/QUgttdfR00Rx-CKkoIkny-vQ/effects-of-backup-frequency-for-continuous-backups]()"

The schedule defines:
* Check the backup is running as it should be, and start it if it isn't;
* Check for retention points that have expired and need to be deleted.

But then I should have only one recovery point in S3 as in RDS.

Ankush_s786 · Answer

In AWS Backup, you can create Amazon S3 backups types that can be continuous or periodic.

- Continuous backups allow you to store backups for up to 35 days and restore to any point in time for the past 35 days.
    - Periodic backups are considered snapshot backups and you can restore only the data that was present at the time of the backup. Periodic backups allow you store backups for longer periods of time to satisfy compliance requirements.

More details here- https://repost.aws/knowledge-center/backup-s3-continuous-periodic-backups

Periodic backups run based on the Backup frequency in your backup plan. S3 Continuous backups takes the backup of the most recent 15 minutes of activities and it is independent of the schedule.  As mentioned in [doc](https://docs.aws.amazon.com/aws-backup/latest/devguide/point-in-time-recovery.html#point-in-time-recovery-considerations), Amazon S3 allows restores up until the most recent 15 minutes of activity.

Since you're unable to see any periodic recovery points in the Vaults so likely only S3 Continuous backups is configured for the s3 bucket.

For example you have a backup plan to take the backup of the s3 bucket with both continuous and periodic backup types with frequency of 1 days & retention period of 35 days. So after end of the month (considering 30 days in a month), there will be 1 s3 continuous backup and 30 s3 periodic backup for the s3 resource inside the vault.

Note: Assuming there is no backup job failure.

Further, In case you see recovery window of 3 days so likely the s3 Continuous backup was created 3 days ago. Please let us know in case you've additional query related to s3 backup.

S3 AWS Backup - Understanding scheduling & incremental & continuous backup

Relevant content