Important
Feature unavailable in Flex Clusters
Flex clusters don't support this feature at this time. To learn more, see Atlas Flex Limitations.
This page describes how to mitigate data loss and execute an easy restore process with Continuous Cloud Backups.
If you need a more granular restore that follows specific criteria, see Restore Specific Documents from Online Archive.
Prerequisites
You must have configured Online Archive on your Atlas cluster to use Continuous Cloud Backups.
Considerations
Your Atlas cluster backup configuration only captures data that is active in your cluster at the time of the backup. However, archived data has the same redundancy guarantees that the S3 vendor provides.
The order in which you execute archive and backup operations has a direct impact on what data you restore to your cluster. Consider the following scenarios:
Scenario 1: You archived data before backing up.
Part of your data lives in your cluster and the remaining part lives in your Online Archive. If your cluster goes down, you restore data to the last state preserved in your backup strategy. Your Online Archive remains unchanged.
Scenario 2: You backed up data before archiving.
You took a snapshot of the entirety of your data before archiving part of it. If your cluster goes down and you restore to a snapshot older than the moment you archived data, you may restore older data that already lives in your Online Archive.
If you use Continuous Cloud Backups, you can replay the oplog to restore your cluster to a point in time that matches the last time documents on the cluster were archived and avoid any data loss or redundancy.
Procedure
If you have configured Continuous Cloud Backups and Online Archive on your Atlas cluster, take the following steps to restore your cluster:
Pause Online Archive.
Restore your cluster from the Continuous Cloud Backups.
Resume Online Archive.