Atlas snapshot-copy regions synchronized with cluster topology
Parent: MongoDB Backup and Restore · Published reference · snapshot 2026-10-27
↓ Facts as markdownall context files
Source-anchored research for Atlas snapshot-copy regions synchronized with cluster topology. Run status: COMPLETED.
These notes link each claim to its source. Some citations identify archived research reports that are no longer hosted on this site. A published reference means the content is available; it does not certify independent review or accuracy.Read the editorial policy and follow the sources before relying on a claim.
Definitions
- Regional placement is the baseline: Atlas documentation places default snapshots in the current primary's provider region, while Manning's chapter describes same-region snapshot storage and preferred-region storage for multi-region clusters. Additional-region distribution requires a copy policy. (medium confidence) (verified as of 2026-10-14) [source]
Structure and components
- At the underlying AWS layer, EBS supports copying completed snapshots to another Region for disaster recovery. This independently establishes a regional-copy primitive, not Atlas's topology reconciliation or its restore-time guarantees. (low confidence) (verified as of 2026-10-27) [source]
How it works
- Atlas's Automatically Sync Copy Regions with Cluster Regions setting creates snapshot-copy policies for secondary regions and updates destinations when secondary regions are added or removed. (high confidence) (verified as of 2026-10-14) [source]
- When a configured snapshot-copy secondary region becomes the primary region, Atlas updates the copy policy to distribute snapshots from the new primary to the former primary and other configured secondary regions. (high confidence) (verified as of 2026-10-14) [source]
- Removing a region from copy policy does not by itself establish immediate deletion of its existing snapshots: Backup Compliance Policy with Keep all snapshots when removing additional snapshot regions retains copies until scheduled expiration. (low confidence) (verified as of 2026-10-14) [source]
- The Terraform documentation warns that, when copy_settings is omitted, removing auto_copy_settings_enabled or setting it false removes the managed copy configuration and deletes the copies it created. (low confidence) (verified as of 2026-10-27) [source]
- When the configured primary region changes to a region already receiving snapshot copies, Atlas updates the copy policy to include the original primary region. The 10 December 2025 changelog documents this behavior separately from secondary-region synchronization. (medium confidence) [source]
Problems, failure modes and limitations
- Terraform's cloud_backup_schedule documentation requires omitting copy_settings when auto_copy_settings_enabled=true and describes a plan-time error if both are configured. Automatic policy entries can appear after apply completes without an operation to wait on. (low confidence) (verified as of 2026-10-27) [source]
- For faster cross-project restores, Atlas additionally requires an eligible target: it cannot be global or multi-cloud, and a multi-region target needs the selected snapshot in every region. Storage capacity, AWS IOPS, NVMe, and older AWS source-cluster enablement also affect eligibility. (low confidence) (verified as of 2026-10-14) [source]
Comparisons and alternatives
- Snapshot copies in every deployed cluster region enable faster direct-attach recovery from local copies instead of slower cross-region streaming after a regional outage; synchronization maintains the destination coverage needed for this approach. (high confidence) (verified as of 2026-10-14) [source]
Facts and statements
- New automatic copy-policy items inherit the cluster backup policy's frequency and retention. Operators can customize these values, but must disable automatic sync before manually adding or deleting copy regions. (low confidence) (verified as of 2026-10-14) [source]
- Enabling automatic region sync overwrites previously configured snapshot copy-policy items; review existing custom policy before enabling it. (low confidence) (verified as of 2026-10-14) [source]
- The Atlas Administration API enables topology synchronization with autoCopySettingsEnabled=true on Update Cloud Backup Schedule for One Cluster. The backup schedule exposes the copy policy through copySettings. (low confidence) (verified as of 2026-10-14) [source]
- Atlas copy destinations must use the cloud provider of the primary region, evaluated per zone for global clusters; topology synchronization does not document unrestricted cross-provider snapshot copying. (low confidence) (verified as of 2026-10-14) [source]
- Continuous Cloud Backup must be enabled before configuring oplog copies for regional point-in-time restore. Enabling oplog copy for one policy item enables point-in-time restore across copy policies in that region. (low confidence) (verified as of 2026-10-14) [source]
- The March 2026 Atlas changelog explicitly announced an automatic copy-region option that follows cluster topology. The September 2022 announcement already described automatic copying to selected regions, so automatic scheduled copying alone does not establish topology reconciliation. (high confidence) [source]
- A practitioner administration notebook independently describes enabling multi-region snapshot distribution for added backup redundancy, but provides no evidence for automatic topology synchronization. (low confidence) [source]
- The API enables synchronization through autoCopySettingsEnabled=true on the Cloud Backup schedule. Enabling it overwrites existing copy-policy items and initially matches the cluster backup policy's frequencies and retention. (low confidence) (verified as of 2026-10-14) [source]
- With synchronization enabled, the Atlas UI/API allow frequency and retention customization but require disabling synchronization before manually adding or deleting copy regions. (low confidence) (verified as of 2026-10-14) [source]
Children
- No children recorded.