Configure Atlas to Automatically Copy Snapshots to Other Regions
On this page
You can configure Atlas to automatically create copies of your snapshots and oplogs and store them in other regions. With snapshots distributed across multiple regions, you can still restore your cluster if the primary region goes down, enhancing your disaster recovery capabilities to bring them in line with internal or regulatory requirements. Additionally, Atlas can perform optimized direct attach restores rather than slower streaming restores for clusters based in the same project and regions where you store copies. This reduces system recovery time.
If you need to restore your cluster, Atlas decides whether to use the original snapshot or a copied snapshot for optimal restore speeds. Atlas might use copied snapshots to optimize restore if you are restoring a cluster in the same region as a snapshot copy, including multi-region clusters if the snapshots were copied to every cluster region. Additionally, if the original snapshot becomes unavailable due to a regional outage within your cloud provider, Atlas uses a copy in the nearest region to enable restores regardless of the cloud region outage.
Note
If you have a Backup Compliance Policy enabled with the Keep all snapshots when removing additional snapshot regions option set to On and you enable, modify, or delete multi-region snapshots, any snapshots already taken remain.
You can use copied oplogs to perform continuous cloud backup from secondary regions in the event the primary region of your cluster goes down.
Note
Serverless instance backups and legacy backups don't support this configuration.
If you enable a peering connection and restrict certain regions for Google Cloud, you can copy snapshots to only those regions.
Snapshots taken in AWS and Azure regions are incremental. Snapshots taken for GCP are not incremental, resulting in higher data storage and transfer costs.
Required Access
To modify or delete backup policies, you must have Project Owner
access to the project. Users with Organization Owner
access must add themselves as a Project Owner
to the project before they can modify or delete backup policies.
Procedure
To configure automatic snapshot distribution in other regions:
Navigate to the Backup Policy tab. In the Additional Backup Copies section, click the arrow button to expand the pane.
Toggle Copy to other regions on. You will see a table of distribution policies pre-populated with the cluster's current regions.
Note
For a global cluster, you can enable or disable this setting independently for each zone.
In the Copy to other regions table, click Choose a region in the Region column of the row under your current region and select the region to distribute snapshots to.
Note
You can select regions only for the same cloud provider as the primary region. For example, if your cluster is in the AWS
us-east-1
region, you can distribute snapshots only to other AWS regions supported by Atlas. For global clusters, each zone may have nodes in multiple cloud providers. When you select regions in which to store copied snapshots, you can select only regions in the same cloud provider as the zone's primary region.In the Snapshots column, click the arrow to view the available snapshot policies for the cluster. Atlas displays options based on the backup policies for the cluster.
For example, if you set the cluster with a Daily backup policy of taking daily 5:00 PM snapshots, checking Daily in the Snapshots column enables snapshot distribution of daily 5:00 PM snapshots. You can't set the snapshot distribution policy to a different frequency or timing than its underlying backup policy.
Note
If overlapping policy items generate the same snapshot, Atlas associates the snapshot with the policy item with the longest retention time.
For example, if the policy speficies a daily snapshot with two-day retention and a weekly snapshot every Saturday with three-week retention, then Atlas displays a frequency of Weekly in the Frequency column on the Snapshots page for the snapshot taken on Saturday and retains the snapshot for three weeks. This is because the weekly policy item has a longer retention time than the daily policy item.
You can enable or disable any combination of the available policies on a region-by-region basis.
Example
The following table describes the snapshot distribution policy for a cluster with enabled storage nodes in the AWS regions
us-east-2
,us-west-1
, andus-west-2
:RegionHourlyDailyWeeklyMonthlyYearlyOn DemandPoint-in-Timeus-east-1
(Original)Every 6 hoursEvery 2 daysEvery SundayLast day of the monthDec. 1stYesYesus-east-2
(Copy)DisabledDisabledEnabledEnabledEnabledDisabledNous-west-1
(Copy)EnabledEnabledEnabledEnabledEnabledEnabledYesus-west-2
(Copy)EnabledEnabledDisabledDisabledDisabledEnabledYesIf you enabled Continuous Cloud Backup for the cluster, you can enable oplog distribution for each additional region. If you disabled Continuous Cloud Backup for the cluster, you can't enable oplog distribution for additional regions.
To delete a policy for snapshot distribution to other regions:
In the Action column, click .
In the dialog box, select either Keep all snapshots in this region or Delete all snapshots in this region, and click Confirm. If you choose to keep snapshots in a disabled region, the snapshots persist until their scheduled expiration date.
Alternatively, to disable all snapshot distribution policies within a single-region cluster or within a global cluster zone, click the Copy to other regions toggle. As with deleting an individual policy, you can choose to keep or delete all copied snapshots for that cluster or zone.
Note
In the event of a cloud provider service outage in a region configured for automatic snapshot distribution, Atlas will attempt restore using snapshots stored in copy regions.