Snapshots and Cloud Manager
A snapshot captures the state of a Managed VM — its disk contents and, optionally, its memory — at a point in time, so you can roll the VM back to that state after a failed update, a configuration change, or a maintenance procedure.
Thinfinity® Cloud Manager does not create or restore snapshots. Those operations happen in the provider’s or hypervisor’s native console, under your own provider session. Provider-side snapshots still interact with Cloud Manager in two ways that matter: the order in which you sequence them with Cloud Manager’s power actions, and how the Instances tab reflects them.
Snapshots are a point-in-time rollback mechanism for an existing VM. They are not Golden Images, which are base images used to provision new instances. See Golden Images: Concept and Role for the distinction.
Sequencing Snapshots with Cloud Manager
Take the snapshot against a quiescent disk. Use Cloud Manager’s Stop action to shut the VM down first, then create the snapshot from the provider console — a live snapshot taken while applications hold open write transactions can capture an inconsistent disk state. See VM Actions Reference for the Stop action.
The credential Cloud Manager uses to connect to the provider does not need snapshot permissions. Snapshots are taken under your own provider session, outside Cloud Manager.
After restoring a snapshot on the provider, use the refresh control at the top of Cloud Manager to force an immediate poll of the provider’s current state rather than waiting for the next polling cycle.
How Cloud Manager Reflects Provider Snapshot Operations
Snapshots are managed on the provider side, but the operations have observable effects on the VM’s state in the Cloud Manager Instances tab.
| Provider operation | Effect in Cloud Manager |
|---|---|
| Creating a snapshot (live) | The VM remains Active. Brief I/O pauses may occur on some providers but do not register as state changes. |
| Creating a snapshot (powered off) | The VM displays as Inactive throughout, following a prior Stop. |
| Restoring a snapshot | Cloud Manager does not observe the transient states of an externally-triggered restore — Activating and Inactivating are generated only for actions Cloud Manager initiates. The instance holds its last-known state until the next poll, then jumps directly to the post-restore state (Active or Inactive). Expect a brief mismatch between the actual VM state and what Cloud Manager displays. |
| Deleting a snapshot | No effect on the VM’s state in Cloud Manager. |
Cloud Manager records only the actions it initiates. Externally-triggered snapshot and restore operations do not appear in the per-instance View Log dialog or the Audit Log — check the provider’s own audit log or activity feed for that history.
Where Snapshots Are Managed
Snapshots are created and managed in your provider’s or hypervisor’s native console. Terminology differs across platforms — Hyper-V uses checkpoints, AWS uses EBS snapshots, OCI uses boot and block volume backups — but the concept is the same. For the procedure on your platform, consult the provider’s documentation.
For provider-level capabilities and limitations, see Supported Providers and Hypervisors.
See Also
- VM Actions Reference — Stop a VM before taking a snapshot
- Golden Images: Concept and Role — Base images for provisioning new instances (distinct from snapshots)
- Supported Providers and Hypervisors — Provider-level capabilities and limitations