Volume snapshots and backups

Volumes (block storage) support two ways to protect their data:

Snapshot: a point-in-time mark kept on the volume’s own storage backend. Created in seconds and cheap, because only blocks that diverge after the snapshot are stored. Good as a safety net before a risky change.
Backup: a full copy of the volume streamed to the backup storage assigned to the volume’s hypervisor (the same destination used for instance backups). Slower and larger, but it survives a total loss of the source storage.

Both are full copies; there are no incremental chains for volumes.

	Snapshot	Backup
Lives on	The volume’s own backend	The hypervisor’s backup storage
Speed	Near-instant	Scales with volume size and network
Survives storage loss	No	Yes
Best for	Pre-change rollback	Retention and disaster recovery

Where to find it

There is no separate admin index for this feature. Open any volume from Storage > Volumes with View, then switch to the Snapshots & Backups tab. Customers see the same tab on their own volumes.

Take a snapshot

Open the volume’s Snapshots & Backups tab.
On the Snapshots sub-tab, click Take Snapshot.
Optionally enter a name. Leave blank for a generated one.
Confirm. The status moves from pending to creating to available.

Create a backup

On the Backups sub-tab, click Create Backup.
Optionally name the backup.
Confirm. The export streams to the hypervisor’s backup storage under a volumes/<volume-id>/ prefix. The status becomes available when the export finishes.

If the hypervisor has no backup storage assigned, the backup fails immediately. Assign one under the hypervisor’s Backups & HA tab, Backup Configuration card; see Instance backups.

Restore

Click the restore icon on any available snapshot or backup. The dialog offers two modes:

Roll back this volume (in-place): overwrite the volume with the snapshot or backup contents.
Restore to a new volume: create a fresh volume from it and leave the source untouched. On the admin panel the new volume uses the source volume’s plan and region.

Safety limits

The panel enforces several guard rails, per volume:

One operation at a time. While a snapshot, backup or restore runs, the volume is busy and further operations are rejected until it finishes.
Cooldown. After an operation finishes, the next operation on the same volume is blocked for 5 minutes. The tab shows a countdown.
Plan caps. The volume’s plan sets Max Snapshots (default 5) and Max Backups (default 10). At the cap, new items are rejected until an old one is deleted.
Per-customer in-flight limit. One customer can run at most 3 volume operations at once across all their volumes.
Live-volume protection. In-place restore is refused while the volume is attached to a running instance.

Deleting a snapshot or backup asks for the item’s name as confirmation, then removes it from the backend or the backup storage.

Billing

Retained snapshots and backups can be metered per GB per month through the volume plan’s Snapshot Credit Value and Backup Credit Value fields (both default to 0, free within the caps). Only items in available status are billed. Charges run inside the hourly Cloud Service billing.

Limits

All operations are manual. Scheduled volume snapshots are not available.
A snapshot of a volume attached to a running instance is crash-consistent, not application-consistent. For a clean application state, stop the instance or detach the volume first.

Common problems

“Volume is busy with operation: …” Another operation is running. Wait for the progress banner on the tab to clear. A job that never finishes is failed automatically and the lock is released.
“Volume is in cooldown. Try again in N seconds.” Wait for the countdown to reach zero.
“Snapshot limit reached” or “Backup limit reached”. Delete an existing item, or move the volume to a plan with a higher cap.
In-place restore is refused. The volume is attached to a running instance. Stop or detach it, or use restore-to-new-volume instead.