Continuous Backups

Cluster Eligibility

This feature is not available for:

Cannot Perform Point-in-Time Restores on MongoDB 4.2 Databases

At this time, you can’t perform a point-in-time restore on Atlas clusters running MongoDB 4.2 using any backup method.

Data Transfer Costs Might Increase

With Continuous Backups enabled, your data transfer costs might increase when you upgrade Atlas clusters from MongoDB 4.0 to 4.2. The continuous backup service that facilitates snapshot creation sends more data over the wire on 4.2.

Consider switching to Cloud Provider Snapshots if you notice a large increase in data transfer after you upgrade your Atlas clusters. Backing up clusters using Cloud Provider Snapshots sends much less data over the wire.

Atlas continuous backups take incremental backups of data in your cluster, ensuring your backups are typically just a few seconds behind the operational system. Atlas continuous backups allow you to restore from stored snapshots or from a selected point in time within the last 24 hours. You can also query a continuous backup snapshot.

For sharded clusters, the backup service temporarily stops the balancer via the mongos in order to insert a marker token into all shards and config servers in the cluster. Atlas takes a snapshot when the marker tokens appear in the backup data.

Collection Limit for Atlas Continuous Backup.

If the total number of collections across all databases in a Atlas cluster meets or exceeds 100,000, Atlas continuous backup can’t take snapshots and may be unable to perform any type of restore for that cluster.

If you need to enable Atlas continuous backup in a project where one or more deployments meets or exceeds the collection limit, click Support in the left-hand navigation bar of the Atlas UI and filling in the requested information to open a support ticket.

Cloud Provider Snapshots do not have a collection limit.

Enable Backup

You can enable continuous backups when you:

If you switch to continuous backups from cloud provider snapshots, Atlas immediately terminates existing backups on your behalf. If you need to retain any of your cloud provider snapshots, download them before switching backup methods. To learn how to download a cloud provider snapshot, see the Restore a Cluster from a Cloud Provider Snapshot page.

From the cluster configuration modal, toggle Do you want to enable backup? to Yes. For clusters deployed to Azure, select the Continuous card.

Enable Point-In-Time Restore for Sharded Clusters

Sharded clusters require enabling cluster checkpoints to support point-in-time restoration from a continuous backup snapshot. Without cluster checkpoints, Atlas can only restore from a snapshot and not from a point-in-time between snapshots. After enabling continuous backups on the cluster, do the following to enable cluster checkpoints:

  1. From the context menu, select the desired organization.
  2. Navigate to the Continuous Backup section of the Atlas UI by clicking Continuous Backup in the Project section of the left side navigation.
  3. Click the ellipsis (...) next to your cluster’s name and select Edit Snapshot Schedule.
  4. Enable the checkbox marked Create cluster checkpoint every: and select an interval from the dropdown menu.
  5. Click Submit to save your changes.


The sharded cluster balancer must pause whenever a cluster checkpoint is created.

Snapshot Storage Location

Each project has one backup data center location dictated by the first backup-enabled cluster created in that project.

For single-region Atlas clusters created after June 21st, 2017, Atlas stores backup data for an Atlas cluster in a data center specific to the cluster’s geographic region.

For multi-region Atlas clusters, Atlas stores backup data for the cluster in a data center specific to the geographical location of the cluster’s Preferred Region.


All additional backup-enabled clusters created in the project use the backup data center selected during deployment of the project’s first backup-enabled cluster regardless of the geographical location of the cluster region or regions.

To change the data center location of a cluster, you must disable backups for all clusters in the project. You can then enable backups for a cluster whose geographic region corresponds to the data center location of your choice.

Cluster Location Backup Service Location
Germany Germany
United Kingdom United Kingdom
Australia Australia
All other cluster locations Ireland

Atlas retains these snapshots based on the retention policy.

Snapshot Schedule

Atlas has the following snapshot schedule, which determines the frequency and the retention of the snapshots:


If you disable backup for a cluster or terminate a cluster, Atlas immediately deletes the backup snapshots for the cluster.

Snapshot Schedule Default Retention Policy Maximum Retention Setting
Base snapshot every <x> hours 2 days
5 days
(30 days if snapshot taken every 24 hours)
Daily snapshot 7 days 360 days
Weekly snapshot 4 weeks 52 weeks
Monthly snapshot 13 months 84 months

By default, Atlas takes base snapshots every 6 hours.

To change a deployment’s snapshot schedule, including the retention policy, go to Backup. For the cluster whose snapshot schedule you wish to modify, click the ellipsis and select Edit Snapshot Schedule menu option.

See also

To modify the schedule using the API, see Snapshot Schedule.