Protection Concepts
The Protection section of the Dashboard covers three things: point-in-time snapshots, external backups, and migrating data into a volume.
Snapshots​
A snapshot captures the state of a volume at a moment in time. Snapshots live inside the cluster and can be used to restore a volume to a prior state, or copied into a brand-new volume.
| Property | Details |
|---|---|
| Scope | Per volume |
| Creation | Manual, on a recurring schedule, or taken automatically as part of a backup |
| Type | Shown as Manual, Scheduled, or Backup so you can tell them apart |
| Restore | Reverts the volume to the snapshot state, replacing current data |
| Copy | Creates a new volume from the snapshot's point-in-time data; original and copy can be deleted independently |
| Retention | Retention in days — an expired snapshot is deleted automatically |
Snapshots taken automatically for a backup do not count against your snapshot limit. If a scheduled snapshot is blocked by the limit, expired snapshots are cleaned up and the snapshot is retried; if it still cannot proceed, a warning appears in System → Logs.
Backups​
A backup exports volume data to a destination outside the volume, providing off-cluster protection.
| Property | Details |
|---|---|
| Scope | Per volume |
| Type | Full — all data; Incremental — only changes since the last backup |
| Target Location | An NFS or filesystem path. S3 targets are not yet supported and are rejected. |
| Restore | Returns the volume to exactly the state it had when the backup was taken |
| Cross-volume restore | A full backup can be restored into a different volume — the basis for disaster recovery onto a fresh volume. An incremental backup restores only to its original volume. |
Re-running a backup, or one that fails partway, cannot corrupt or overwrite the previous good copy. A restore from external storage is refused if the copy was altered.
Scheduled backups with automatic expiry of older copies are supported by the REST API
(schedule and retentionPolicy). The Dashboard covers on-demand create, trigger, and cancel;
your most recent backup is always kept.
Migrations​
A migration copies data from an existing source into a MASS volume — useful when moving off another storage system. While a migration is in flight the target volume becomes read-only and is shielded from deletion, so the in-progress copy cannot be clobbered.
| Source | Dashboard | CLI (mass-ctl migration) |
|---|---|---|
| NFS | Supported | Supported |
| Local (hostPath) | Supported | Supported |
| S3 | Not yet supported | Supported |
Choosing Between Them​
- Snapshots — fast, frequent recovery points inside the cluster
- Backups — off-cluster durability and disaster recovery
- Migrations — one-time or staged import of data from outside MASS