This version of the documentation is archived and no longer supported. To learn how to upgrade your version of MongoDB Ops Manager, refer to the upgrade documentation.
You were redirected from a different version of the documentation. Click here to go back.

Backup Preparations

Before backing up your cluster or replica set, decide how to back up the data and what data to back up. This page describes items you must consider before starting a backup.


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.

For an overview of how Backup works, see Backup.

Backup Configuration Options

The backup and recovery requirements of a given system vary to meet the cost, performance and data protection standards the system’s owner sets.

Ops Manager Enterprise Backup and Recovery supports five backup architectures, each with its own strengths and trade-offs. Consider which architecture meets the data protection requirements for your deployment before configuring and deploying your backup architecture.


Consider a system whose requirements include low operational costs. The system’s owners may have strict limits on what they can spend on storage for their backup and recovery configuration. They may accept a longer recovery time as a result.

Conversely, consider a system whose requirements include a low Recovery Time Objective. The system’s owners tolerate greater storage costs if it results in a backup and recovery configuration that fulfills the recovery requirements.

Ops Manager Enterprise Backup and recovery supports the following backup architectures:

  • A File System on a Sophisticated SAN
  • A File System on one or more NAS devices
  • An AWS S3 Blockstore
  • MongoDB Blockstore in a Highly Available configuration
  • MongoDB Blockstore in a Standalone configuration


The backup architecture features and concerns are provided as guidance for developing your own data protection requirements. They do not cover every scenario nor are they representative of every deployment.

Backup Method Features

Backup System Feature File System on SAN File System on NAS AWS S3 Blockstore MongoDB HA Blockstore MongoDB Blockstore
Snapshot Types Complete Complete Many partial Many partial Many partial
Backup Data Deduplication If SAN supports No Yes Yes Yes
Backup Data Compression Yes Depends Yes Yes Yes
Backup Data Replication If SAN supports No No Yes No
Backup Storage Cost Higher Medium Lower Higher Lower
Staff Time to Manage Backups Medium Medium Lower Higher Medium
Backup RTO Lower Medium Lower Lower Medium

When Do You Use a Particular Backup Method?

  • If you do not want to maintain separate backup systems nor do you want your staff to maintain them, consider backing up to a MongoDB or S3 blockstore.
  • If you need to restore data without relying on MongoDB database, consider backing up to a file system on a SAN or NAS device or an S3 blockstore.
  • If you are backing up large amounts of data or frequently need to restore data, consider either a file system on a SAN, S3 blockstore or a MongoDB blockstore configured as a replica set or sharded cluster.
  • If you want to minimize internal storage and maintenance costs, consider backing up to a MongoDB standalone blockstore or an S3 blockstore.
  • If you have a SAN with advanced features like high availability, compression, deduplication, etc., consider using that SAN for file system backups.

Backup Sizing Recommendation

When sizing the backup of your data, MongoDB recommends keeping replica set size to 2 TB or less of uncompressed data. If your data increases beyond 2 TB, you should shard that database and keep each shard to 2 TB or less of uncompressed data.

These size recommendations are a best practice, and are not due to a limitation of the MongoDB database or Ops Manager.

Backup and restore can use great amounts of CPU, memory, storage and network bandwidth.


Your stated network throughput is a theoretical maximum. That value doesn’t account for sharing or throttling of network traffic.

Consider the following scenario:

  • You want to back up a 2 TB database.
  • Your hosts support a 10 Gbps TCP connection from Ops Manager to its backup storage.
  • The network connection has very low packet loss and a low round trip delay time.

A full backup of your data would take more than 30 hours to complete. [*]

This doesn’t account for disk read and write speeds, which can be, at most, 3 Gbps reads and 1 Gbps writes for a single or mirrored NVMe storage device.

If you shard this database into 4 shards, each shard runs its backup separately. This results in a backup that takes less than 8 hours to complete.

[*]These throughput figures were calculated using the Network Throughput Calculator.

Snapshot Frequency and Retention Policy

By default, Ops Manager takes a base snapshot of your data every 24 hours.

If desired, administrators can change the frequency of base snapshots to 6, 8, 12, or 24 hours. Ops Manager creates snapshots automatically on a schedule; you cannot take snapshots on demand.

Ops Manager retains snapshots for the time periods listed in the following table. If you terminate a backup, Ops Manager immediately deletes the backup’s snapshots.

Snapshot Default Retention Policy Maximum Retention Policy
Base snapshot 2 days 5 days (30 days if frequency is 24 hours)
Daily snapshot 0 days 1 year
Weekly snapshot 2 weeks 1 year
Monthly snapshot 1 month 7 years

You can change a backed-up deployment’s schedule through its Edit Snapshot Schedule menu option, available through the Backup page. Administrators can change snapshot frequency and retention through the snapshotSchedule resource in the API.

Changing the reference time changes the time of the next scheduled snapshot:

  • If the new reference time is before the current reference time, the next snapshot occurs at the new reference time tomorrow. See the first two rows of the table below for examples.
  • If the new reference time is after the current reference time, and you make the change before the current reference time, the next snapshot occurs at the new reference time today. See the third row of the table below for an example.
  • If the new reference time is after the current reference time, but you make the change after the current reference time, the next snapshot occurs at the new reference time tomorrow. See the fourth row of the table below for an example.
Time of Change Current Reference Time New Reference Time Time of Next Snapshot
08:00 UTC 12:00 UTC 10:00 UTC 10:00 UTC tomorrow
13:00 UTC 12:00 UTC 10:00 UTC 10:00 UTC tomorrow
08:00 UTC 12:00 UTC 14:00 UTC 14:00 UTC today
13:00 UTC 12:00 UTC 14:00 UTC 14:00 UTC tomorrow

If you change the schedule to save fewer snapshots, Ops Manager does not delete existing snapshots to conform to the new schedule. To delete unneeded snapshots, see Delete a Snapshot.

Namespaces Filter

The namespaces filter lets you specify which databases and collections to back up. You create either a Blacklist of those to exclude or a Whitelist of those to include. You make your selections when starting a backup and can later edit them as needed. If you change the filter in a way that adds data to your backup, a resync is required.

Use the blacklist to prevent backup of collections that contain logging data, caches, or other ephemeral data. Excluding these kinds of databases and collections will allow you to reduce backup time and costs. Using a blacklist is often preferable to using a whitelist as a whitelist requires you to intentionally opt in to every namespace you want backed up.

Storage Engine

To backup MongoDB clusters, use the WiredTiger storage engine storage engine.

If your current backing databases use MMAPv1, upgrade to WiredTiger:

MongoDB removed the MMAPv1 storage engine in MongoDB 4.2. To learn more about storage engines, see Storage in the MongoDB manual.


  • Ops Manager does not backup deployments where the total number of collections on the deployment meets or exceeds 100,000.
  • Ops Manager does not replicate index collection options.


Changed in version 3.4: Starting in 3.4, Ops Manager supports encryption for any backup job that was stored in a head database running MongoDB Enterprise 3.4 or later with the WiredTiger storage engine.

For details on setting up backup encryption, see Encrypted Backup Snapshots.

Resyncing Production Deployments

For production deployments, it is recommended that as a best practice you periodically (annually) resync all backed-up replica sets. When you resync, data is read from a secondary in each replica set. During resync, no new snapshots are generated.

You may also want to resync your backup after:

  • A reduction in data size, such that the size on disk of Ops Manager’s copy of the data is also reduced. This scenario also includes if you:
  • A switch in storage engines, if you want Ops Manager to provide snapshots in the new storage engine format.
  • A manual build of an index on a replica set in a rolling fashion (as per Build Indexes on Replica Sets in the MongoDB manual).


For sharded clusters, checkpoints provide additional restore points between snapshots. With checkpoints enabled, Ops Manager creates restoration points at configurable intervals of every 15, 30 or 60 minutes between snapshots. To enable checkpoints, see enable checkpoints.

To create a checkpoint, Ops Manager stops the balancer and inserts a token into the oplog of each shard and config server in the cluster. These checkpoint tokens are lightweight and do not have a consequential impact on performance or disk use.

Backup does not require checkpoints, and they are disabled by default.

Restoring from a checkpoint requires Ops Manager to apply the oplog of each shard and config server to the last snapshot captured before the checkpoint. Restoration from a checkpoint takes longer than restoration from a snapshot.

Snapshots when Agent Cannot Stop Balancer

For sharded clusters, Ops Manager disables the balancer before taking a cluster snapshot. In certain situations, such as a long migration or no running mongos, Ops Manager tries to disable the balancer but cannot. In such cases, Ops Manager will continue to take cluster snapshots but will flag the snapshots with a warning that data may be incomplete and/or inconsistent. Cluster snapshots taken during an active balancing operation run the risk of data loss or orphaned data.

Snapshots when Agent Cannot Contact a mongod

For sharded clusters, if the Backup Agent cannot reach a mongod process, whether a shard or config server, then the agent cannot insert a synchronization oplog token. If this happens, Ops Manager does not create the snapshot and displays a warning message.