Cloud Manager Overview


MongoDB Cloud Manager is a service for managing, monitoring and backing up a MongoDB infrastructure. In addition, Cloud Manager allows Administrators to maintain a server pool to facilitate the deployment of MongoDB.

Cloud Manager provides the services described here.


Cloud Manager Monitoring provides real-time reporting, visualization, and alerting on key database and hardware indicators.

How it Works: A lightweight Monitoring Agent runs within your infrastructure and collects statistics from the nodes in your MongoDB deployment. The agent transmits database statistics back to Cloud Manager to provide real-time reporting. You can set alerts on indicators you choose.


Cloud Manager Automation provides an interface for provisioning machines, configuring MongoDB nodes and clusters, and upgrading your MongoDB deployment.

"Automation coordinates MongoDB instances running in a public cloud, in your private data center, or on your local system."

How it Works: Automation Agents on each server maintain your deployments. The Automation Agent also maintains the Monitoring and Backup agents and starts, restarts, and upgrades the agents as needed.

Automation allows only one agent of each type per machine and will remove additional agents. For example, when maintaining Backup Agents, automation will remove a Backup Agent from a machine that has two Backup Agents.


Cloud Manager Backup provides scheduled snapshots and point-in-time recovery of your MongoDB replica sets and sharded clusters.

How it Works: A lightweight Backup Agent runs within your infrastructure and backs up data from the MongoDB processes you have specified.


Only sharded clusters or replica sets can be backed up. To back up a standalone mongod process, you must first convert it to a single-member replica set.

Data Backup

When you start Backup for a MongoDB deployment, the agent performs an initial sync of the deployment’s data as if it were creating a new, “invisible” member of a replica set. For a sharded cluster the agent performs a sync of each shard’s primary and of each config server. The agent ships initial sync and oplog data over HTTPS back to Cloud Manager.

The Backup Agent then tails each replica set’s oplog to continually update the backup. The backup is consistent with the original primary up to the last oplog supplied by the agent.

Backup performs the initial sync and the tailing of the oplog using standard MongoDB queries. The production replica set is not aware of the copy of the backup data.

Backup uses a mongod with a version equal to or greater than the version of the replica set it backs up.

Backup takes and stores snapshots based on a user-defined snapshot retention policy. Sharded clusters snapshots temporarily stop the balancer via the mongos so that they can insert a marker token into all shards and config servers in the cluster. Cloud Manager takes a snapshot when the marker tokens appear in the backup data.

The procedures used to reduce the storage space required to store a snapshot depend on where it is stored, as described in the following table:

Snapshot Store Description
MongoDB blockstore Each successive snapshot stores only the differences from the previous snapshot. Compression and block-level de-duplication reduce the size of snapshot data.
AWS S3 bucket Each successive snapshot stores only the differences from the previous snapshot. Compression and block-level de-duplication reduce the size of snapshot data.
File system store Depending on the configuration, compression reduces the size of the snapshot data.

All snapshots represent a full backup. See Backup Configuration Options for more information about backup configurations.

Data Restoration

Cloud Manager Backup lets you restore data from a scheduled snapshot or from a selected point between snapshots. For sharded clusters you can restore from checkpoints between snapshots. For replica sets, you can restore from selected points in time.

When you restore from a snapshot, Cloud Manager reads directly from the snapshot storage and transfers files either through an HTTPS download link or by sending them via HTTPS or SCP.

When you restore from a checkpoint or point in time, Cloud Manager first creates a local restore of a snapshot from the snapshot storage and then applies stored oplogs until the specified point is reached. Cloud Manager delivers the backup via the same HTTPS or SCP mechanisms. To enable checkpoints, see Enable Cluster Checkpoints.