Cross-region snapshots let you keep a copy of a Cloud VM’s recovery point in a region other than where the source VM lives. The primary use case is disaster recovery — if a region is unavailable, you can spin up a new VM in another region from the copied recovery point.
When to use this
- Disaster recovery (DR) — keep a recent recovery point off-region so a regional incident doesn’t take your data with it.
- Region migration — move a VM from one region to another by copying a snapshot, restoring as a new VM, then decommissioning the original.
- Latency-driven cloning — stand up a clone closer to a different user base.
For routine, in-region protection, regular snapshots and scheduled backups are sufficient.
Available regions
Cloud VMs are currently available in the Asia region across these locations:
A multi-region (cross-continent) layout is on the roadmap. Until then, “cross-region” snapshot copy applies between locations within Asia.
Workflow
The high-level pattern, regardless of which UI controls are surfaced for copy:
11. Take a snapshot in the source region
In the source VM’s Snapshots tab, create a recovery point with the Capture Mode that matches your DR plan (Create a snapshot). For DR, All attached volumes is the safest default.
22. Copy to the target region
From the recovery point’s Details view, copy the snapshot to the target region. The platform replicates root and captured data disks to the destination region’s storage and produces a recovery point there.
The copy operation runs as an activity event you can track via the Activity feed.
33. Restore in the target region
In the target region, open the copied recovery point and use Create New VM to provision a new VM from it. Pick a plan available in the target region.
44. Cut over
Update DNS, callers, or load balancer backends to point at the new VM. Verify, then decommission the original VM if you’re migrating rather than running active-passive DR.
What’s copied
Cross-region snapshots replicate the same content as in-region snapshots:
- The root disk.
- Any captured data volumes (per the Capture Mode used at snapshot time).
Mount instructions for data volumes remain available on the copied recovery point — they’re identical to the source because device paths inside the guest don’t depend on region.
Considerations
- Transfer time scales with recovery storage. A 60 GB recovery point will take longer to copy than a 10 GB one.
- Cost — cross-region snapshots are billed for storage at both ends until you delete one. See Backup pricing; cross-region copy may have an additional transfer component.
- Eventual consistency — copying is asynchronous; don’t restore from the destination until the copy event reports succeeded.
- Plans available in the target region — the Available Upgrades dropdown only shows plans the target region offers. Pick a plan equal to or larger than the source VM’s plan to avoid disk shrink issues.
DR drill checklist
Test your DR plan periodically — a snapshot you’ve never restored is just hope.
- Take a fresh snapshot in the source region.
- Copy it to the target region.
- Restore as a new VM in the target region.
- Verify the application boots, the data is intact, and external dependencies (DNS, DBs, queues) are reachable.
- Document any guest-level steps (mount commands, service start order) that aren’t part of the platform restore.
- Delete the test VM and copied recovery point if you don’t need to keep them.
Related pages
